Systems and methods for nucleic acid sequencing

SYSTEMS AND METHODS FOR NUCLEIC ACID SEQUENCING

STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH AND DEVELOPMENT

Aspects of this invention were made with government support under a Qualifying Therapeutic Discovery Grant awarded by the 1RS, in conjunction with the Department of Health and Human Services. The U.S. government may have certain rights in the invention.

BACKGROUND OF THE INVENTION

Methods for quick and cost effective genetic and biological analysis, including high-throughput DNA sequencing, remain an important aspect of advancing personalized medicine and diagnostic testing. Current high throughout or miniaturized systems have limitations. For example, current systems for DNA sequencing, including those that employ optical detection, are cumbersome and expensive, and have limited throughput. While some systems use sensors and sequencing flow cells to address these limitations, these are generally one-time use disposables, which substantially increases the cost to the user and limits the complexity of the sensor, since the sensor must be cost effectively manufactured for a single use. Emulsion PCR provides some advantages, however sequencing clonal DNA populations can exhibit limited accuracy when sequencing docs not proceed “in phase” throughout the clonal population, which in-turn can lead to, in effect, short read lengths.

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

A need exists for systems and methods for genetic and biological analysis, and in particular, methods and systems for highly parallel or clonal sequencing reactions that are both sensitive and cost effective.

BRIEF SUMMARY OF THE INVENTION

The aspects and embodiments described herein relate to systems and methods for sequencing polynucleotides, as well as detecting reactions and binding events involving other biological molecules. The systems and methods may employ chamberfree devices and/or nanosensors to detect and/or characterize such reactions in highthroughput. Because the system in many embodiments is reusable, the system can be subject to more sophisticated and improved engineering, as compared to single use devices.

In some embodiments, the invention provides methods and systems for sequencing polynucleotides, which may be individual double or single stranded polynucleotides, or in other embodiments are clonal populations of polynucleotides. For example, one aspect of the invention provides a method for parallel or clonal polynucleotide sequencing, the method comprising: sequencing a first portion of a population of target polynucleotides, correcting for phase error, and then sequencing a second downstream portion of the population of target polynucleotides. In various embodiments, the polynucleotide sequencing may involve one or more of: clonal sequencing of a bead array, electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate reaction components.

For example, phase error may be corrected by adding a combination of three nucleotide bases to halt the population of polynucleotides at the first occurrence of the excluded base. Phase error may also be corrected through the combination and/or order of incorporation reactions as described in detail herein. Alternatively or in addition, phase error may be corrected by reversibly incorporating, into the in-phase polynucleotide strands, a chain terminating nucleotide. Alternatively or in addition, phase error may be corrected by adding one or more oligonucleotide clamps, the clamp(s) hybridizing to the target polynucleotides to halt the sequencing reaction. In some embodiments, the clamp is denatured, destabilized, or degraded to continue the sequencing reaction. In other embodiments, at least one clamp has a 3’ terminating nucleotide that cannot be extended, and thus upon removal of the 3’ terminating nucleotide, the clamp becomes a primer for subsequent downstream sequencing.

Re-phasing can occur at regular intervals, or alternatively, the reaction can be monitored for loss of signal, and rephasing conducted to restore sequencing signal.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In another aspect, the invention provides a method for reducing leading phase error in parallel or clonal polynucleotide sequencing. The method according to this aspect comprises sequencing a population of target polynucleotides in the presence of a competitive reaction, where the competitive reaction comprises nucleotide bases or nucleotide derivatives for all four nucleotide bases. Generally, three of the four nucleotide bases will be unincorporable into the growing polynucleotide strand, thereby decreasing the propensity of the polymerase to incorporate incorrect nucleotides. According to this aspect, the polynucleotide sequencing may optionally involve one or more of: clonal sequencing of a bead array, electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate reaction components. Various nucleotide derivatives are known and described herein which may be bound by the polymerase, but not incorporated into the growing polynucleotide strand.

In still other aspects, the invention provides a method for reducing lagging phase error in parallel or clonal polynucleotide sequencing reactions. In accordance with this aspect, the method comprises stockpiling polymerase enzyme on or near a population of target polynucleotides during a sequencing reaction, such that polymerase is substantially available for each active polymerization site. Alternatively or in addition, the method comprises binding a repair protein or single stranded DNA binding protein to the population of target polynucleotides. Optionally, the polynucleotide sequencing reaction involves one or more of: clonal sequencing of a bead array, electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate reaction components. The stockpiling can be a result of the native binding of the polymerase to primers, including non-extendable primers hybridized to the template polynucleotides.

In still other aspects, the invention provides methods for repeated nucleotide sequencing, such that several sequencing runs can be analyzed for sequence data. According to this aspect, the method comprises providing a circularized DNA sequencing template, and sequencing the template by determining the sequence of incorporation of nucleotides by a DNA polymerase having 5’ to 3’ exonuclease activity. This aspect may also optionally involve one or more of: clonal sequencing of a bead array, electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate reaction components. The DNA polymerase according to this aspect may be highly processive and have reduced exonuclease activity. The highly processive polymerase may be bound on or near a biosensor adapted to measure the incorporation of nucleotides.

In some embodiments, the method sequences a single DNA molecule by attaching a polymerase enzyme to a biosensor in a volume and causing a DNA template with associated primers to enter the volume and hybridize and be held by (or in

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 proximity to) the polymerase. Subsequently, the sequence may be determined upon extension of the primers by the polymerase.

In yet other aspects, the invention provides a chamber-free device, comprising: an electromagnetic sensor array, a magnetic carrier for carrying or holding molecules of interest at or near the electromagnetic sensors, and a mechanism for removing the magnetic carrier via liquid flow and/or electromagnetic removal. The electromagnetic sensor may be one of a nanoneedle or a nanobridge, and the device may further comprise local amplifiers. In some embodiments, the electromagnetic sensor has a narrow structure, and is etched under the structure such that both sides of the sensor’s surface arc accessible to changes in pH, or to changes in conductivity. The devices and methods described herein may include one or more improvements including incorporating materials having a reduced zeta potential, using reagents that allow for both polynucleotide incorporation and sensitive pH measurements, and design and fabrication of nanosensors in optimal proximity and configuration relative to a bead or substrate holding polynucleotide templates for a sequencing reaction.

Other aspects and embodiments of the invention will be evident to one of skill in the art based on the following detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1A shows a complete integrated system, along with schematic details of some subsystems. Figure IB shows schematic details of a sample and library' prep subsystem. Figure IC shows schematic details of a DNA fragmentation and purification subsystem. Figure ID shows a PDMS library preparation subsystem.

Figure 2 shows a magnetic and virtual confinement array.

Figure 3 shows a Comsol simulation the electric fields for a virtual well.

Figures 4 A-4D show a schematic, drawings and a fabricated PDMS valving subsystems.

Figure 5 shows an embodiment for a combined NanoNeedle sensor and magnetic array element.

Figure 6 shows two versions of magnetic arrays which may position magnetic beads in fixed locations.

Figure 7 shows an embodiment of a magnetic array which may locate beads in a fixed location.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Figure 8 depicts schematically an element of an array utilizing a “leaky valve” to localize a bead.

Figure 9 depicts a simulated voltage and current plot associated with a redox reaction for detection.

Figure 10 illustrates the charge distribution of a DNA binding protein.

Figure 11 shows a schematic depiction of a magnetic array utilized for planar magnetic particles.

Figure 12 schematically illustrates a combined magnetic, virtual well, and NanoNeedle array element with a bead.

Figures 13 A-13E show drawings, illustrations, and photomicrographs of various enrichment module embodiments.

Figure 14 illustrates the bead loading density for an existing flow cell.

Figure 15 schematically illustrates the valving system and interfaces for a multichannel flow cell with proximate valving control.

Figure 16 schematically illustrates a simple model for the impedances in a NanoNeedle and bead array element.

Figure 17 schematically depicts an under-etched stacked NanoNeedle.

Figure 18 is a photomicrograph of an array of under-etched stacked NanoNeedles.

Figure 19 schematically depicts a 2D array of under-etched stacked NanoNeedles.

Figure 20 is a photomicrograph of a 2D array of under-etched stacked NanoNccdlcs.

Figures 21 A-C schematically depicts an element of an array of a single side contact NanoNeedles.

Figures 22 A-D schematically and diagrammatically show an element of an array of a double sided contact NanoNeedle.

Figure 23 schematically depicts the elements of a NanoBridge.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Figures 24 A-C schematically depicts views of a ring NanoBridge.

Figure 25 schematically depicts am array of NanoNeedles utilized for single molecule sequencing.

Figure 26 illustrates sequencing data and the linearity of same from a NanoNeedle array element.

DETAILED DESCRIPTION OF THE INVENTION

The present invention provides for methods and systems for DNA sequencing, and other types of biological or genetic analysis. The invention provides methods and systems for sequencing clonal DNA populations or arrays of single molecule DNA, including by electronic sequencing, thereby providing a low cost and convenient sequencing platform. In some aspects, the invention provides methods that monitor for and/or correct for phase error during sequencing clonal populations of DNA, to thereby improve accuracy and read lengths. Alternatively, the invention provides methods and sensors for sequencing single molecules of DNA, to thereby avoid such phase errors. In other aspects, the invention provides arrays, including magnetic arrays, and virtual reactors for highly parallel reactions. These systems in some embodiments include nanosensors for detecting biological reactions or interactions, including incorporation of nucleotides during DNA sequencing. Further, the invention provides integrated systems for amplifying and sequencing DNA samples.

Monitoring and Correcting for Sequencing Phase Errors

As used herein, “phase error” is defined as the occurrence where some template polynucleotides of a clonal population are extended more or less than the consensus state. For fragments where a base is added where it shouldn’t be added relative to the consensus, this phase error is considered to be “leading.” For other template molecules where a base is not added where it should be added relative to consensus, the polynucleotide is considered to be “lagging.” Since polymerases are imperfect, some phase error is inevitable within a colony that has a long extension reaction as a part of a colony based sequencing process. Phase error limits the read lengths of commercial clonal sequencing systems.

“Leading sequencing incorporation error” refers to sequences that get ahead of the dominant sequence through incorrect additions of nucleotides. The incorrect additions may result from polymerase errors, particularly when high concentrations of dNTPs are used in a noncompetitive reaction. Alternatively, the leading sequencing incorporation error may result from inadequate washing or nonspecific binding of dNTPs, which may be subsequently released and incorporated. “Lagging sequencing

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 incorporation errors” refers to sequences that get behind the dominant sequence through missed additions of the correct nucleotide; this may occur due to non-optimal reaction conditions, steric hindrance, secondary structure, or other sources of polymerase inhibition. Longer cycle times can allow more opportunities for the polymerase to incorporate the wrong nucleotide. Similarly less accessible DNA may result in inadequate opportunities to incorporate the correct nucleotide. It is anticipated that temperature, step times, polymerase selection, nucleotide concentration, salt concentration and buffer selection may be optimized to minimize incorporation errors.

For example, a DNA sample may have a sequence of TGTTC in a first region after a region which is complementary to a primer. A fluidic cycle may first introduce dCTP, secondly followed by dTTP, thirdly followed by dATP, and fourthly followed by dGTP, interspersed with wash steps. In the first part of a fluidic cycle, dCTP molecules which flow in as part of said first cycle may not be properly washed out of a well structure. In a second part of a fluidic cycle, dTTP molecules which flow in as part of said second cycle may not be properly washed out a well structure. During the first and second part of the first fluidic cycle, no dNTPs should be incorporated. During a third part of a fluidic cycle, dATPs may be introduced and may be incorporated, as dATP is complementary to T, the first base of the sample. Any nonspecifically bound dCTP molecules which cease to be nonspecifically bound may also be incorporated during this third portion of a fluidic cycle. These unbound dCTP molecules may be incorporated after a dATP molecule is incorporated. After a dCTP molecule is incorporated, two more dATP molecules may subsequently be incorporated, which may result in some of the molecules of a monoclonal bead having leading sequencing phase errors. Thus some molecules of a monoclonal bead may become “out of phase”.

When a polymerase is provided with a single nucleotide or nucleotide analog at a time, the error rate is typically significantly higher than when all four nucleotides or nucleotide analogs are provided. This may occur despite the enormous difference in the catalytic efficiency measured as kpop'Kd.app, which may be four logs or more lower for a mismatched nucleotide vs. a matched nucleotide. Most of this is due to the difference in Kd.ap_P. For example, Klenow polymerase has a misincorporation rate of one base in z O every 10 to 10 bases. In comparison, polymerase extension reactions by current commercial systems utilizing the incorporation of single native dNTPs may be limited to 100 to 1000 bases. The polymerase in these systems spends almost all of its time trying to misincorporate bases, leading to significant “leading” phase errors. Alternatively dephasing may result from a polymerase not incorporating a base in an incorporation fluid flow cycle due to the absence of said polymerase, followed by the presence of a polymerase in a subsequent incorporation fluid flow cycle, or from a sufficiently low combination of dNTP concentration and time for incorporation such that a base is not incorporated, resulting in “lagging” phase error. Even when a 7

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 nucleotide added to the system is the next nucleotide to be added, the reaction time must be long enough to complete the reaction for a homopolymer, which may be eight or more nucleotides, or for DNA strands that may be less accessible from steric hindrance.

In one aspect, the invention provides methods for parallel or clonal polynucleotide sequencing. In certain embodiments the method comprises sequencing a first portion of a template polynucleotide population, and correcting for phase error. Sequencing then continues to a second downstream portion of the target polynucleotide population. In various embodiments, the sequencing may involve one or more of clonal sequencing of an array of polynucleotide populations (e.g., a bead array), electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate sequencing reaction components. In various embodiments, the invention provides methods for monitoring for and correction both leading and lagging phases, and the various approaches described herein may be used individually or in any combination.

As used herein, “clonal” means that substantially all of the populations of a bead or particle may be of the same template nucleic acid sequence. In some embodiments there may be two populations associated with a single sample DNA fragment, as would be desired for “mate pairs,” “paired ends”, or other similar methodologies; the populations may be present in roughly similar numbers on the bead or particle, and may be randomly distributed over the bead or particle.

In some embodiments, the colony is re-phased by providing sequencing by incorporation nucleotides in different orders than might be otherwise normally done. For example, if a system predominately has lagging phase error (as opposed to leading phase error), with for example a simple 1% lagging error per base (and all four different bases have similar lagging error rate), after 20 bases have been incorporated, just over 75% of the members of the colony may be in phase, while over 20% may be lagging by a single base. By the time 70 bases have been sequenced, less than half of the members of the colony will be in phase, 35% will be lagging by a single base, 13% will be lagging by two bases, and 3% will be lagging by three bases. So, for the following exemplary incorporation sequence example where the ideal position is shown in bold (...CGATCGATCGA), 50% of the colony will be in phase at the fourth base (T), 35% will be lagging one base (A), 13% will be lagging two bases (G), and 3% will be lagging three bases (C). If the previous order for incorporation of the bases had been CGAT, and a C is provided, the lagging error will continue, and will be slightly increased. If instead C is excluded the next base provided is G, the leading base will not be extended, while the portion of the colony which is lagging two bases at the first G shown w'ill be extended; if an A is provided next, most of the colony will now be in phase. If a three base combination without C, for example GAT is provided one or more times, any phase error will be concentrated at C bases. Statistically some

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 sequences may become more out of phase as a result, but most sequences may be made to be more in phase. Two base combinations may also be used in much the same manner to re-phase the colonies, and a mixture of two base and three base sets may be used. After rc-phasing, the system may revert to four base combinations and rephasing can be repeated as frequently as necessary.

In other embodiments, four base combinations may not be used at all, but alternating three and two base sets are used exclusively. In a further embodiment, the four bases are added in any combination of three and two base sets of nucleotides, with the composition of the two and three base sets alternating in some embodiments. In some embodiments, said base sets described may also include the use of unincorporablc nucleotides. In other embodiments, the concentrations of the nucleotides and or unincorporable nucleotides utilized in two, three or four base combinations may vary from cycle to cycle, or from set to set.

In certain embodiments, phase error is corrected by excluding at least one nucleotide base from a sequencing reaction. For example, phase error can be corrected by adding a combination of three nucleotide bases, thereby pausing each nascent polynucleotide in the clonal population at the first occurrence of the excluded nucleotide base.

In certain other embodiments, phase error is corrected by reversibly incorporating, into the in-phase polynucleotide strand, a chain terminating nucleotide. Once lagging phase strands have caught up to the in-phase strand, the terminating nucleotide is removed. This approach may be most advantageous when the sequence being sequenced includes a homopolymer region. For example in the following sequence fragment ....AGCTCCC, where the in phase portion of the colony has incorporated the T base, with most of the lagging sequence having incorporated the C, G and A bases as the final bases of the members of the colony, if a C’ terminating nucleotide is provided, followed by the base combination AGT, AGT, then there may be a predominantly bimodal population, where the sequences ....AGC’ and ....AGCTC’ predominate. Said terminator may then be removed from the C’ nucleotides, and another C’ terminating nucleotide may be provided, resulting in two predominant sequences: ....AGC and AGCTCC’. The C’ terminated nucleotide may then be followed by the base combination AGT, AGT, resulting in the two populations: ....AGCT and ....AGCTCC’. The terminator may then be removed, and nonterminated C nucleotides may then be provided, resulting predominantly in a single sequence: ....AGCTCCC.

In some embodiments, phase error is anticipated at certain positions (e.g., homopolymeric regions) based on a reference sequence, thus allowing the phase correcting to be efficiently implemented at an appropriate place in sequencing.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

These approaches may be most effective for those systems which have a predominate source of error, such as, for example, a lagging error. Base combinations which may be used for re-phasing may be added individually, so that the complete sequence of incorporation may be determined, or may be added together, so that rephasing may be accomplished with a small section of missing data. In some embodiments, reversible terminators may be used repeatedly during a sequencing process or method, and may be combined with incorporable or unincorporable nucleotides.

In yet other embodiments, phase error is corrected by adding one or more oligonucleotide clamps, the clamps hybridizing to the target polynucleotide to halt the sequencing reaction, and thereby remove phase error. Such a clamp could be a PNA fragment, a DNA fragment, or other molecule, native or non-native, which binds specifically to a sequence of DNA. In a system which is utilized for targeted resequencing, specific oligos may be used as “clamps”. The “clamps” may be provided at the same time that primer sequences may be provided, prior to when primer sequences may be provided, after primer sequences may be provided, before any sequencing reactions have been completed, or after some sequencing reactions have been completed. Multiple different targeted or untargeted clamps may be provided for each template.

Said clamp(s) may be random or targeted to specific regions of a DNA template. A DNA fragment or other clamp may be further stabilized by the use of histones, cationic protamines, recombinase, and other molecules known to stabilize duplex DNA. Sequence reactions may then proceed up to the point of the clamp(s). Additional incorporation reactions may be performed, using single bases, two base combination, three base combinations, or four bases simultaneously. Said clamps may be positioned such that said clamps may be spaced such that (on average) from ten to fifty bases exists between the 3’ end of the primer and said clamp(s), or may be positioned such that (on average) ten to 100 bases exists between the 3’ end of the primer and said clamp(s), or may be positioned such that (on average) 100 to 500 bases may exist between the 3’ end of the primer and said clamp(s), or may be positioned such that (on average) 300 to 500 bases may exist between the 3’ end of the primer and said clamp(s), or may be positioned such that (on average) 1000 to 5000 bases or more may exist between the 3’ end of the primer and said clamp(s), or may be positioned such that (on average) 2000 to 5000 bases may exist between the 3’ end of the primer and said clamp(s).

In some embodiments, the clamp may have a specific number of bases which may specifically hybridize, and may have additional bases which may serve to stabilize said clamp. If the sequence of said clamp is not targeted to a specific region(s), but is instead a non-targeted clamp, the sequence of the clamp may be selected using several criteria, including the stability of the damp, the frequency of the selected clamp sequence in the genome of interest, or in genomes of a similar nature, or in a chromosome of interest, or in a transcriptome of interest. The hybridization stability of the complete clamp, including any non-specific bases such as deoxyinocinc, 5nitroindolc, or abasic nucleotides, or may include any of the universal bases described in US 7,575,902, and may include the stability of the bases selected as specific bases for the clamp.

In some embodiments, said clamps may comprise 5, 6, 7, 8, 9, 10 or more specific bases. Said clamps may be used for a number of DNA colonies, wherein substantially all of the colonies may have different DNA sequences from other DNA colonies. In some embodiments a single clamp type, comprising a single set of specifically hybridizing dNTPs may be used. In other embodiments, multiple clamp types, wherein the number or order or spacing of specifically hybridizing bases may be different. For example, two different hexamer clamps may be used to decrease the average spacing, as measured in DNA bases, from the primer to the clamp, or between one clamp and the next clamp, that which would occur if only one of the two hexamer clamps were utilized, but may be larger than that which might occur were a single pentameric clamp to be used. In some embodiments, the spacing, as measured in DNA bases from the primer to clamp, or from clamp to clamp may be varied as a result of the choice of the sequence of the clamp, as there is significant variation (more than 20X) in the representation of different hexamers in the transcriptome (Anderson et al RNA VI 4(5)).

In some embodiments, the clamp(s) are subsequently removed (after phasing) by raising the temperature, changing the pH or ionic concentration, resulting in the denaturation of the clamp, but leaving the longer extended primers, which may subsequently be further extended after the removal of the clamp(s). In other embodiments, the clamp(s) may comprise a nick site(s) which may be subsequently nicked by an appropriate nickase or endonuclease, which may destabilize the clamp sufficiently to denature it. In some embodiments, the clamp may comprise clcavablc linker sites, where said cleavable linker site(s) may be chemically cleavable or photocleavablc. In some embodiments, a base terminated at the 3’ position may be provided as a part of the clamp(s). Said terminators may be subsequently removed after nucleotides have been added so as to effect rephasing. Said terminator may be removed using chemical or photochemical processes. In some embodiments, a combination of different types of cleavable linker sites (e.g., unique cleaablc linker sites) arc used for different clamps, so that the clamps with different linker types may be provided prior to beginning any sequencing, or after sequencing has commenced, and different clcavablc mechanism may be used to denature the clamps in an order which permits multiple rephasing of the template DNA.

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments, the method comprises adding clamps after a sequencing and rephasing process has occurred. In further embodiments, the process of adding additional clamps may be repeated multiple times, such as 2 to 5 times, 4 to 10 times, or more than 10 times.

In some embodiments, data may be collected as incorporations may be performed for all bases preceding a position adjacent to said clamp; in other embodiments, data may not be performed for all bases preceding a position adjacent to said clamp.

In some embodiments, said clamps may be utilized in combination with nonstrand displacing polymerases, such that when said polymerase reaches said clamp through a polymerization process, said polymerase cannot displace said clamp. In further embodiments, the 5’ base of said clamp may be linked utilizing a non native linker which cannot be cleaved by a 5’ to 3’ exonuclease activity which said polymerase may have. In an alternative embodiment, said polymerase may be a non-strand displacing polymerase and may further be lacking 5’ to 3’ exonuclease activity.

In a further embodiment, a strand displacing polymerase may be used in combination with a clamp which is resistant to strand displacement by said strand displacing polymerase. Said clamp may consist of, particularly at the 5’ end of the clamp, non native bases that arc resistant to the strand displacement activity of a strand displacing polymerase. Such a base may comprise an abasic base, such as a base which has been depurinated, or synthetic bases such as PNAs, arabinosyl derivatives of nucleobases, ribonucleotides, 2’-O-alkylribonucleotides, 2’-O-methylribonucleotides, or bases with methylphosphonate linkages.

In some embodiments, the clamp, after phasing, is used as a primer. Said clamp may include a reversible terminator at its 3’ terminus, where the primer extension reaction proceeds until the clamp substantially prevents further extension. Further extension may be followed be the removal of the terminator from the 3’ terminus of the clamp, permitting a polymerase to initiate a primer extension reaction from said clamp/primer.

In some embodiments, a single clamp/primer is used for a colony or a set of colonies wherein the distance between said primer and said clamp may be significantly more than the average sequencing length before dephasing normally would occur, for example, when it is desirable to use said clamp(s) for the purpose of determining the structure of the DNA e.g. creating a scaffold and removing sequence ambiguity due to repetitive sequences. In some embodiments, the distance between the primer and the clamp/primer may be twice as long as the average sequence “read length” before dephasing, or may be from twice as long to five times as long as the average sequence

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 “read length” before dephasing, or may be from five times as long to twenty times as long as the average sequence “read length” before dephasing, or may be from twenty times as long to fifty times as long as the average sequence “read length” before dephasing. Additional clamp/primcr(s) may optionally be utilized in this embodiment to extend the read length, and or elaborate further the structure of the DNA colony(s). In some embodiments, the average read length is about 50 nucleotides, or about 100 nucleotides, or about 200 nucleotides, or about 300 nucleotides, or about 400 nucleotides, or about 500 nucleotides. In other embodiments, a clamp/primer is used for a colony or a set of colonies where the distance between said primer and said clamp/primer or from the clamp/primer to the next clamp/primer may be similar to the average sequencing length before dephasing normally would occur, for example, as would be desirable to extend the length of read of the average sequencing length.

If there is any variation in the stopping point of incorporation as a result of interactions between the clamp, including any stabilizing moiety, and the polymerase (or ligase), a clamp rc-phasing method may be combined with one of the methods previously described, which may be advantageous as the sequence of the clamp is already known, permitting addition of bases other than the first base of the clamp sequence, potentially followed by bases other than the second base of the clamp sequence, or any stopping at any other known portion of the clamp sequence.

In order to permit short hybridization probes as rephasing reagents, stabilizing compounds such as hydralazine, or antitumor antibiotic cc-1065 may be employed. Similarly the probe may be a PNA or LNA probe, which may provide the dual function of providing tighter binding, and precluding the need to prevent the probe from being extended by polymerase, by using for example, a terminator at the 3’ end of the probe. Additionally the probe may be a single plex, a duplex which may hybridize to the target DNA to create a more stable triplex, or may be a triplex, which may hybridize with the target DNA to form a quadraplex. In some embodiments the probe may be provided in two or more pieces, wherein one portion may be a hybridizing single plex, and a second portion may be hybridize to create a triplex. In some embodiments additional portions to the probe complex may be provided, allowing the formation of a quadraplex, or the formation of a duplex with more than two pieces in addition to the original template.

In other embodiments, lagging dephasing may be reduced by “stockpiling” polymerase enzymes on or near the DNA which is to be extended and sequenced, such that a number of polymerases may be available for each active polymerization site. Said stockpiling may result from native binding of the polymerase to the DNA. Such binding may result normally, as for example when a Klenow polymerase is used, where the Klenow polymerase has intrinsic ssDNA and dsDNA binding.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Alternatively, in some embodiments, the DNA may be provided with 3’ terminated random primers in addition to universal or targeted primers, where said universal or targeted primers may be not terminated, and where said polymerase may bind at the 3’ terminated end of said random primers, as well as to the 3’ end of said universal or targeted primers. As said random primers may be terminated, said random primers may not be extended, and thus may not contribute to the signal concomitant to extending the strand from said universal or targeted primers. In this embodiment, the polymerase may be lacking in 3’ to 5’ exonuclease activity, such that said random primers may be not degraded, resulting in a loss of stockpiling capacity.

In an alternative embodiment, random primers using nucleotide analogs in at least the 3’ terminus may be employed instead of 3’ terminated random primers, where polymerase will bind to the random primers, but will not extend them. In this embodiment, the polymerase may have 3’ to 5’ exonuclease activity if said 3’ to 5’ exonuclease activity is effectively inactive in removing the nucleotide analogs. In some embodiments, it may be desirable for the Ka to be smaller for a polymerase binding to said 3’ terminated random primers, or random primers having one or more nucleotide analogs in the at least 3’ terminus position. Said nucleotide analog containing random primers may be chimeric, where said chimera comprises native nucleotides and nucleotide analogs, multiple types of nucleotide analogs, or native nucleotides and multiple types of nucleotide analogs.

In a further embodiment, the random primers may be 3’ terminated random primers, wherein the 3’ terminus of the random primers further comprises a thiophosphate nucleotide in the 3’ (terminated) position, such that the random primers are further resistant to 3’ to 5’ exonuclease activity. The 3’ thiophosphate primers may be commercially available from, for example, IDT (Integrated DNA Technologies). In this embodiment, native polymerases with 3’ to 5’ exonuclease activity such as phi29 may be used, without needing to mutate the polymerase to inactivate the exonuclease activity to prevent degradation of said random primers. Such thiophosphates may be alpha-S or alpha-R steroisomers. The random primers may also comprise 5’ thiophosphates, such that 5’ to 3’ exonuclease activity may be inhibited. Alternatively, the random primers comprise 3’ inverted dT, which may act to prevent both polymerization and exonuclease activity with respect to the 3’ position of the random primer. Dideoxynucleotides may be used as terminators. Said terminators may be reversible terminators, virtual terminators, terminators attached to the base of the nucleotides, or to any position of the sugar the nucleotides. The nucleotides in the random primers may be natural bases, or may be a synthetic bases. Said random primers may comprise dNTPs, or may comprise chimeras in combination with PNA, RNA, LNA, 5-Nitroindole, deoxyinosine, or other non-natural dNTPs.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Stockpiled polymerases may also be bound to the surface of a bead, to the surface of a sensor, to interstitial regions between sensors, to groups, linkers, or polymers attached to said beads, sensors or interstitial regions. Such binding may be to additional strands of non extendable exonuclease resistant DNA, synthetic DNA or other linear polymers, or may be other binding groups such as antibodies, wherein the binding groups may bind directly to said polymer, or may bind to an intermediate protein which may complex with said polymerase.

The relationship in the relative kinetics between the K<,ff of the polymerase from the active incorporation site, and the Kofi' stockpile site(s) and the number of stockpiled polymerases, and the extension period must be appropriate in order to insure that a stockpiled polymerase will be able to replace a polymerase which has disassociated with the active incorporation site of a DNA strand in order to incorporate a base(s) and prevent lagging dephasing within a desired eiror rate. For example, if the Koff is the same for both the active incorporation site, and the stockpile site(s) K_off and K_on, and the Koff is equivalent to 20 incorporation fluidic cycles, if one had 20 stockpiled polymerases, the odds that another polymerase will become available to bind to the active incorporation site is less than 50%. This can be improved by reducing the K_off of the stockpile sites, and increasing the Ko_n of the Stockpile sites, with the caveat that loss of polymerase to fluid flow may be an issue. The Koff may be reduced as a result of utilizing non natural bases, terminators, or the association of proteins which may normally reduce the processivity of the polymerase to the binding site.

Alternatively or in addition, the polymerase may be a single type of polymerase, or may be a combination of different types of polymerases. In general, commercialized more processive polymerases have lower incorporation error rates, and as such it may be desirable to mainly use highly processive polymerases. It may be desired to have one type of polymerase have a significantly longer than another type of polymerase.

It may also be appropriate to have more shorter Koff polymerase available to replace any more highly processive (longer Koff) polymerase that disassociates, such that a polymerase will be available to incorporate any bases as appropriate. As such, it may be appropriate for the Koff of the less processive to be less than the time period allocated for an incorporation fluidic cycle, in order to insure that polymerases will be available for incorporation should a more processive polymerase become disassociated from the binding site of the extending primer(s).

In some embodiments where it is preferred to use two or more types of polymerase, it may be desirable to preferentially bind a more processive polymerase to the primer which is to be extended. It may thus be desirable to allow' a more processive polymerase to bind to the primer which is to be extended prior to adding a less processive polymerase, which may have ssDNA and or dsDNA binding moieties

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 associated therewith. In other embodiments, it may be desirable to modify or mutate a polymerase such that a binding moiety may be added to said polymerase, such that the polymerase may bind directly to portions of ssDNA or dsDNA.

The stockpile of polymerases may be replenished periodically. Said replenishment may occur with every incorporation cycle, or after an appropriate number of incorporation cycles have occurred. The number of cycles between replenishment may vary depending on the stockpiling method. For example, if the stockpiling method is storage on random primers, the number of stockpile locations reduces with the extension of the universal or targeted primer, as said random primers may be displaced by the polymerase, using cither strand displacement, or 5’ to 3’ exonuclease digestion of the random primers. The number of stockpiled polymerases may be maintained, if for example, a second binding mechanism exists for binding to double stranded DNA.

In certain embodiments, the method comprises monitoring the reaction for loss of signal, and rephasing to restore sequencing signal. For example, data from a sequencing reaction may be monitored, and rephasing may be performed when it is observed to be needed, as for example, as seen by signal levels which may be less than expected for a single base, but would be expected if lagging phase error were present, or by the reduction of the signal level observed for single bases. Said observations may consider nominal sequence in determining whether a signal would statistically result from lagging phase error. Said observations may include a histogram of the signal levels for a sequencing fluidic cycle or a set of sequencing fluidic cycles.

In some embodiments rephasing may be performed for any clonal sequencing system, including those which utilize four incorporable nucleotides, as well as all of those described above with respect to minimizing dephasing. In some embodiments, the rephasing is performed in connection with emulsion PCR, or alternatively, the magnetic or bead array described herein, optionally in connection with electronic sequencing.

In one embodiment of the current invention, compensation may be performed to reduce for phase error by using earlier and/or later data to determine expected background levels for each cycle for each location. Expected phase error for each base, for each base in the sequence context, and the amount of lagging and leading error previously determined may be used to assist in determining the actual base. This error correction may also take into account phase errors from neighboring reactions on an array, as well as the influence of neighbors on the signals received from each sensor.

In some embodiments the distribution between leading and lagging phase error is influenced such that one type of phase error may occur at a higher rate than the other type of phase error. In one embodiment the concentration of dNTPS may be limited, so

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 that lagging phase error is more likely than leading phase error. In a further embodiment, rephasing may then be performed to correct for the more probable phase error type. In other embodiments, the method acts to correct for both types of phase errors, wherein the method corrects for one phase error type, and then corrects for the other type of phase error. Said method(s) for phase correction may be repeated periodically through the sequence process. In some embodiments, the fluidic pattern for rephasing may be fixed. For example, the fixed pattern may have a fixed number of fluidic sequencing cycles between performing rephasing methods, or the number of fluidic sequencing cycles may change during the sequencing process, for example, by reducing the number of fluidic sequencing cycles between performing rephasing methods.

In another aspect, the invention provides a method for reducing leading phase error in parallel or clonal polynucleotide sequencing. The method according to this aspect comprises sequencing a population of polynucleotides in the presence of a competitive reaction. The competitive reaction comprises either nucleotide bases or nucleotide derivatives for all four nucleotide bases, wherein three of the four nucleotide bases are unincorporable into the growing polynucleotide strand. The sequencing may involve one or more of clonal sequencing of an array of polynucleotide populations (e.g., a bead array), electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate sequencing reaction components. The unincorporable nucleotide in various embodiments may be selected from (without limitation) a PNA nucleotide, a LNA nucleotide, a ribonucleotide, an adenine monophosphate, an adenine diphosphate, an adenosine, a deoxyadenosine, a guanine monophosphate, a guanine diphosphate guanosine, a deoxyguanosine, a thymine monophosphate, a thymine diphosphate 5-Methyluridine, a thymidine, a cytosine monophosphate, a cytosine diphosphate cytidine, a deoxycytidine, a uracil monophosphate, a uracil diphosphate, a uridine, and a deoxyuridin. Generally, the unincorporable nucleotides are bound by the polymerase, but are not incorporated into the growing polynucleotide strand by the polymerase. In some embodiments the concentration of the unincorporable nucleotides is relative to the polymerase activity for each of the unincorporable nucleotides. Said unincorporable nucleotides or nucleotide analogs may be unlabeled, optically labeled, or charge labeled.

In some embodiments, the concentration of the incorporable dNTPs may be increased relative to the concentrations of incorporable dNTPs, permitting a decrease in both the leading and lagging error rates.

In still another aspect, the invention provides a method for reducing lagging phase error in a population of polynucleotide templates. The method comprises stockpiling polymerase enzyme on or near the target polynucleotide during a

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 sequencing reaction, such that polymerase is available for each or substantially each active polymerization site.

Alternatively or in addition, the method comprises binding a repair protein or single stranded DNA binding protein to the target polynucleotide to, among other things, remove secondary structure or aid processivity. In some embodiments, it may be desired to reduce secondary structure of single stranded DNA which may otherwise interfere with polymerase activity, resulting in lagging phase error. In one embodiment, a protein which binds to single stranded DNA is used. Such proteins may include repair proteins such as bacterial RecA DNA repair proteins, HIV nucleocapsid proteins, T4 bacteriophage gene product 32, calf thymus UP1, Epstcin-Barr virus BALF2, or commercialized single stranded binding proteins such as Epicentre E. coli single stranded binding protein, as well as many others. In some embodiments, said single stranded binding proteins may also serve to aid the processivity of the polymerase. In other embodiments other moieties may also aid in the processivity of the polymerase, such as Epstcin-Barr virus BMRF1, triplex S. Ccrcvisiac proliferating cell nuclear antigen, T4 bateriophage gene product 45, thieoredoxin, E. coli PolIII holoenzyme, eukaryotic clamp protein PCNA, or other DNA sliding clamp proteins, or other double stranded or single stranded DNA binding moieties.

In some embodiments, polymerase processivity may be enhanced by mutations to the polymerase, such as the addition of a helix-hairpin-helix domain to Phi29 polymerase, or the use of a chimerical polymerase as described by Salas et al in PNAS 107 16506, the addition of a thioredoxin binding domain, the addition of an archaeal sliding clamp, DNA binding protein Sso7d, a zinc finger domain, a leucine zipper, amongst other possibilities. In some embodiments, the polymerase may be further modified such that more than one processivity enhancement mutation is used, such as mutations to add both double stranded and single stranded binding moieties at the respective ends of the polymerase. In some embodiments, such binding moieties are also indirectly bound to the polymerase to, for example, add a streptavidin moiety by mutating a polymerase, and adding via mutation a biotin moiety to a single stranded DNA binding moiety and or double stranded DNA binding moiety such as those mentioned above. The polymerase will thus be bound to the single or double stranded DNA binding moiety through the streptavidin biotin binding, and further bound to the DNA through said single or double stranded binding moiety. In other embodiments, other binding moieties are employed to bind a polymerase to a single stranded or double stranded DNA binding moiety, wherein an additional moiety may be added via mutation to each of the polymerase and the single or double stranded binding moiety, wherein the additional moieties added to the polymerase and the single or double stranded binding moiety have a mutual binding affinity. In further embodiments, a moiety may be added via mutation to one of the polymerase, a single stranded DNA 18

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 binding moiety and a double stranded DNA binding moiety, wherein the moiety added via mutation may then bind to another of a polymerase, a single stranded DNA binding moiety and a double stranded DNA binding moiety, such that a polymerase is bound to a single or double stranded DNA binding moiety.

In some embodiments, the stockpiling is achieved by the use of the polymerase’s binding to non-cxtcndablc primers. In certain embodiments, the nonextendable primers are not subject to 3’ exonuclease activity of the polymerase. The non-exendable primers in some embodiments are 3’ terminated random primers, and the extension primers are universal or targeted primers. A polymerase binds at the 3’ terminated end of the random primers, as well as to the 3’ end of the universal or targeted primers. For example, the 3’ terminated primers may comprise a thiophosphate nucleotide in the 3’ terminated position, such that said 3’ terminated primers are resistant to 3’ to 5’ exonuclease activity.

The sequencing according to this aspect may involve one or more of clonal sequencing of an array of polynucleotide populations (e.g., a bead array), electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate sequencing reaction components.

In connection with electronic sequencing, in some embodiments it may be desirable to use ion concentrations that arc lower than might be optimal for synthesis, in order to have an ion concentration sufficiently low for improved operation of the detector. This may result in phasing errors, and a shorter sequence length than desired. It may be desirable to have sequence lengths longer than possible with said low ion concentrations. Thus, in one embodiment, the effective read length is increased by alternating conditions optimal for detection, with conditions optimal for synthesis. For example, the method may comprise performing a sequencing reaction to the full length possible while using low ionic concentrations needed for optimal reading of said DNA extension reaction, melting off the extended primer strand, introducing new primers and dNTPs, and proceeding with the synthesis reaction while using optimal ion concentrations for synthesis. The process of melting off the extended primer strand, introducing new primers and dNTPs, and proceeding with the synthesis reaction while using optimal ion concentrations for synthesis, followed by changing the conditions to those appropriate for detection may be repeated multiple times, until the process no longer results in useful data. As the determination of how many synthesis steps to use may be statistical, the process may be reversed, performing a synthesis with conditions optimal for synthesis, followed by performing synthesis using conditions appropriate for detection; this may then be followed by melting off the extended primer strands, introducing new primers, and using ionic concentrations appropriate for detection.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In order to optimize detection sensitivity, ion and or dNTP concentrations may be desirable which are below the concentrations which may be desirable for proper enzyme kinetics. This may result in longer incorporation times than desired, and/or more lagging phase error. In some embodiments, it may be desirable to use more than one concentration of dNTPs during a single incorporation cycle. For example, it may be desirable to perform single incorporation cycles with low concentrations of dNTPs and/or ions so as to optimize the sensitivity and signal to noise of the sensors. Thus, a solution with a low concentration of dNTPs and/or ions may be flowed into a flow cell(s) so that measurements may be taken. This may be immediately followed by a solution with a concentration of dNTPs and/or ions which is optimal for incorporation into the extended primer, such that minimal de-phasing may occur. In an alternative embodiment, a high concentration of dNTPs may be flowed into a flow cell(s), so that a quick and optimal incorporation reaction may occur, providing for minimal dc-phasing. This may then be quickly followed by a reagent solution with low or no dNTPs and an appropriately low ion concentration such that an optimal sensor reading(s) may be taken.

In another embodiment, a reagent solution with optimal concentrations of dNTPs and/or ions for quick incorporation of nucleotides may be flowed into a flow cell(s) to quickly incorporate a significant portion of the dNTPs into the various colonies as needed, followed very quickly by flowing into a flow cell(s) a reagent solution optimal for optimal sensor reading(s), followed by flowing into a flow cell(s) a reagent solution with concentrations of dNTPs and/or ions at concentrations appropriate for incorporation of nucleotides into the extended primers of the various primers. The first reagent solution may have a concentration of nucleotides which may be sufficiently high that were the reagent solution to be used for an extended time period, polymerase incorporation errors might occur at a noticeable level.

Because polymerase efficiencies may be different for each base, it may be desirable to use different concentrations of different dNTPs, use different buffers, different cation concentrations, different concentrations of polymerase, different types of polymerase, or any combination thereof to optimize the incorporation rate, minimize the amount of phase error, minimize the amount of incorporation error, and maximize the read length.

In certain embodiments, the concentration levels of the different nucleotides or unincorporable nucleotide analogs is matched to the relative polymerase activity for each of the nucleotide or nucleotide analogs. For example dTTP binding rate has been measured to be different by a factor of over two with respect to the other nucleotides. The other three nucleotides may be much closer in their polymerase binding rates, but still vary by over 10 percent with respect to each other. It is likely that the difference

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 may be even larger in comparing the polymerase binding rates for different unincorporable nucleotide analogs relative to native nucleotides.

In a further embodiment, the concentrations of the unincorporable nucleotides needed for equivalent polymerase extension are higher, equal, or lower than the concentration for optimal primer extension with minimal dephasing for the one or more incorporable nucleotides or nucleotide analogs which may be provided for a sequencing reaction not utilizing unincorporable nucleotides or nucleotide analogs, making the probability of misincorporation of nucleotides or nucleotide analogs lower than if the concentrations of unincorporable nucleotides analogs were provided such that polymerase extension efficiency were matched, or if the unincorporable nucleotide analogs were provided at concentrations with lower polymerase extension efficiency relative to the incorporable nucleotides or nucleotide analogs. Alternatively, the unincorporable nucleotide analogs may be provided at concentrations with lower polymerase binding rates relative to the incorporable nucleotides or nucleotide analogs, such that the reaction may proceed at a higher rate than would occur if the polymerase binding rates of the unincorporable nucleotides were the same or higher than the incorporable nucleotide analog binding rates.

In other embodiments, the concentration of the incorporable nucleotide or nucleotides provided may be varied such that the incorporation rates for the different dNTPs may be more equal. The concentrations may be modified as needed for different buffer conditions, pHs, polymerases, and interactions with any polymerase clamp complexes, or other moieties which may be utilized to stabilize the polymerase.

In some embodiments, unincorporable nucleotides are used as part of a reagent set which is not intended to cause the incorporation of nucleobases. Such a reagent set may, for example, be intended to wash out a previous set of incorporable nucleobases, prior to introduction of a new set of incorporable nucleobases. Such a wash step may also include phosphatases to degrade triphosphate nucleobases to diphosphates or monophosphates, so that any remaining triphosphates will be degraded and will thus be unincorporable. The unincorporable nucleobases may occupy a position complementary to the next base in the pocket of the polymerase, and may serve to effectively increase the processivity of the polymerase, as the thumb of the polymerase will remain closed, attempting to incorporate the unincorporable nuclcobasc, and reducing the disassociation of said polymerase from the DNA.

The above approaches may be used for reaction conditions where there may be three unincorporable nucleotide analogs and one incorporable nucleotide or nucleotide analog, or where there are two unincorporable nucleotide analogs, and two incorporable nucleotides or nucleotide analogs, or where there may be one unincorporable nucleotide analog, and three incorporable nucleotide or nucleotide analog.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Detection methods which may be used with the above reaction conditions and unincorporable nucleotide analogs may include any form of electronic sensing of incorporation or incorporation events, including ISFETs, CHEMFETs, NanoNeedles, NanoBridges, chcmilumcncsccncc detection, fluorescence detection, including detection of Qdots or other nonstandard fluorophores, and detection of intercalating fluorophores, detection using fluogenic moieties.

In further embodiments of the current invention employing bead arrays and electronic sequencing (as described herein), null beads or null sensor regions (sensor regions which do not have associated colonies), may be used as references; such references may compensate for variations in temperature, variations in the conductivity or pH of the bulk reagent, or localized variations in conductivity or pH. The control of the system will help limit and identify phase errors, thereby extending read length.

A common practice for FET pH sensors is to use a reference electrode; some designs for arrays of FET pH sensors use a reference channel for each detection channel; others have reference channels for a set of detection channels. But the local pH of the detector is influenced by the presence of the DNA colony, and changes as the length of the second strand of DNA is extended by the polymerization reaction. In using a chemistry whereby a single type of nucleotide is introduced to the flow cell at a time, many detector channels will not have a reaction taking place at that detector; in fact most detector channels will not have a reaction occurring. Thus in one embodiment, neighboring detectors are used as reference channels, providing the data analysis algorithms an opportunity to measure the pH or ion concentration as it changes in detectors which are neighboring detectors to a detector which has a polymerization reaction occurring. This permits detection of pH or ion concentration levels, or other sources of noise local to the detector of interest, and may also permit detection of crosstalk, allowing monitoring and modification of the crosstalk deconvolution function.

In certain embodiments, null beads (which do not have DNA colonies or which have colonies that do not have the same primer as that which will be used for the sequencing by synthesis reaction) are used to insure that some detectors will not have a polymerization reaction occurring at them. Beads which have colonies of DNA with an appropriate primer may be introduced to the flow cell, taking a random set of positions on detectors. Subsequently, a set of null beads as previously described may be introduced into the flow cell, whereby the null beads can occupy random locations not already occupied by beads already present in the flow cell. As reagents are introduced into the flow cell, the null beads are then used to monitor the pH and or ion concentration levels, enabling the analysis algorithms to better determine background levels and/or a crosstalk deconvolution function.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In another embodiment, null beads are introduced pair wise with sample beads, and the signals are determined using a differential amplifier, obviating the need for the analysis algorithms to directly deconvolve variations in background and crosstalk.

In still other embodiments, adjacent beads or sensor regions which do not have an incorporation reaction in the current fluid cycle may be used as a reference. As on average most beads or sensor regions will not have an incorporation reaction when sequencing large populations of different sequences, these locations without an incorporation reaction may be used as additional references. As references, beads or sensor regions without an incorporation reaction may also provide better references relative to empty sensors or null beads, as DNA polymerase, and beads will be present in the volume of interest, and any variation in surface chemistry and resulting background counter ion concentration will likely be better matched. It is likely that different colonies on beads or sensor regions may have colony DNA and/or extended primers of different lengths from the lengths of colony DNA and/or extended primers associated with other beads and or sensor regions, and thus may have different amounts of charge present which may interact with the sensor.

Thus in some embodiments of the current invention, software may need to compensate for the relative length of DNA and/or extended primers associated with sensors and the resultant different charge and signal levels, as well as the location of the beads or colony relative to the sensor, which may influence. The software may keep a record of the signal levels: associated with each sensor prior to introduction of a bead to said sensor; after introduction of a bead but prior to introduction of a primer and or a polymerase(s); after introduction of a primer and or a polymerase(s) but prior to introduction of a first nucleotide in a sequencing by synthesis reaction; associated with colonies without primers; associated with colonies which may have different lengths of DNA; and the signal levels associated with colonies with hybridized primers, which may have different lengths of DNA. The software may keep track of how many bases have been added to each primer. The signal levels may be an absolute level, or a relative level between different sensors. The software may use a number of other adjacent, or proximate sensors as references to determine the signal level for an individual sensor, compensating for the length of the DNA for each set of colonies, the length of the extended primers, and the signal level which may be generated as a result of positioning of the beads or colonies relative to the sensor. The software may also compensate for the relative position of the sensors relative to the transition between a reagent volume which docs not have dNTPs and a reagent volume which docs have dNTPs. The software may further compensate for variations in dNTP concentration as a function of diffusion and or dNTP concentration depletion, as a result of incorporation of into extended primers. The amount of diffusion may be characterized from earlier data from the same chip, which may be in the same sequencing run, or from a previous 23

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 sequencing run, or from sequencing data from a previous chip utilized on the same instrument, or from data from other chips utilized on other instruments. The expected level of depletion may be determined based on data generated as the transition between a reagent volume which docs not have dNTPs and a reagent volume which docs have dNTPs moves through a flow cell.

In some embodiments, control beads arc used which have a known sequence, or with different known sequences. The known sequence(s) may have homopolymer runs of different known length, and may be used to calibrate the response of the system to better determine the length of homopolymer runs of unknown length. The known sequences may also be useful in determining the level of signal or background signal, as the length of the extended primer and whether an incorporation event has occurred will be known in advance. The control beads may be used to differentiate instrument problems from sample prep problems. The control beads with known sequences may be generated outside of the system, and introduced with colonies of DNA attached thereon, or control DNA may be mixed with or introduced prior to or after the sample DNA to generate DNA colonies on beads or otherwise associated with the sensors. In some embodiments, signal levels may be monitored and stored in a manner similar to that described herein for use with normalization from adjacent beads.

In a manner similar to optical aberration, the diffusion of species being detected by the detectors will cause crosstalk between different detectors. In one embodiment of the current invention, deconvolution of data taken from different sensors on the array may be performed in a manner similar to that used to deconvolve the point spread function from an optical system. The deconvolution function used may depend in part on the temperature of the flow cell at the time of detection, as well as the flow rate through the flow cell, which may tend to cause more crosstalk “downstream” of a particular colony. The deconvolution function which is utilized for said deconvolution may be a fixed deconvolution function, or may be derived as part of a best fit algorithm.

The sensor array may be made self calibrating, allowing calibration for such variables as amplification efficiency, bead size and loading, bead placement on the sensor, etcetera. In general, in amplifying a DNA sample to create a monoclonal population of DNA on a bead, a first primer may be ligated to the sample DNA prior to said amplification reaction. In the sequencing process, a second primer is provided which is complementary to said first primer which has been ligated to said DNA sample. Said second primer may be several bases shorter than said first primer. Thus each monoclonal bead has a known initial sequence at a density independent of amplification efficiency, which will be the section of said first primer which is not matched by said complementary second primer. This may permit prior knowledge as to

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 the base sequence, and may include calibration sequences such as known length homopolymer runs..

In an alternative embodiment, such calibration may occur after a sequencing reaction is complete, or alternatively after a number of bases have been sequenced in a sequencing reaction. Statistically, most fluidic cycles in a sequencing reaction will not result in a base incorporation at an individual bead. The next most common result from a fluidic cycle will be the incorporation of a single base. Thus the data set may be analyzed, and appropriate signal levels may be set for each bead.

In a further embodiment, ongoing compensation/calibration may be implemented as the signal level for a base incorporation in a fluidic cycle reduces during a sequencing process, and the background for a not having a base incorporation in a fluidic cycle as a result of factors such as loss of some of the clonal population on the bead, sequencing phase lead, or sequencing phase lag of some of the clonal population on the bead or other factors. Thus a signal level may calculated as to how much signal may be expected for a fluidic cycle which has a single base incorporation, no base incorporation, or multiple bases incorporated due to a homopolymer run, at each point in a sequencing process.

In some embodiments, reverse phase alignment may be performed, wherein a polymerase with 3’ to 5’ exonuclease activity is used with a dNTP pool that is missing at least one dNTP. The polymerase with 3’ to 5’ exonuclease activity will remove bases back to the next dNTP in the provided dNTP pool, at which point equilibrium will be reached, and no further nucleotides will be removed. This may be performed in order to remove any bases which have been incorporated due to leading phase error. In other embodiments, one or more base type may be incorporated which does not permit exonuclease activity, such as a thiophosphate nucleotide. Exonuclease activity may then be used by the removal of unincorporable nucleotides, improving the kinetics for exonuclease activity. In other embodiments, initial incorporations may performed with an exo- polymerase, followed by the use a exo+ polymerase or another nuclease to remove any bases back to a thiophosphate nucleotide or other nucleotide which is resistant to nuclease activity.

Single and/or Repeated Polynucleotide Sequencing

In another aspect, the invention provides a method for repeated and/or single polynucleotide sequencing. The method in this aspect comprises providing a circularized DNA sequencing template, and sequencing the template by determining the sequence of incorporation of nucleotides by a DNA polymerase having 5’ to 3’ exonuclease activity. The sequencing according to this aspect may involve one or more of clonal sequencing of an array of polynucleotide populations (e.g., a bead array),

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 electronic detection of nucleotide incorporation, and an electronic well to isolate or concentrate sequencing reaction components. In various embodiments, the DNA polymerase is highly processive and has reduced exonuclease activity. Further, the DNA polymerase may be bound on or near a biosensor adapted to measure the incorporation of nucleotides. In some embodiments, the method comprises pre-binding a polymerase to the polynucleotide prior to sequencing. According to this aspect, the invention avoids the need to correct phasing or re-phasing.

A single DNA molecule can be sequenced by a NanoNeedle biosensor (which is described in detail herein). A polymerase enzyme is attached to the sensor. A DNA sample with associated primers may then be caused to enter the volume with the polymerase attached sensors, using for example, pressure induced flow, electro-osmotic induced flow and or migration, or similar means. A single molecule from the DNA sample may then be bound by a polymerase attached to a sensor in a sensor array. Additional single DNA molecules may also be bound by other polymerases bound to other sensors in the sensor array.

In order to permit repeated measurements of the same DNA sample, the DNA sample may be circularized, and the polymerase may be a strand displacing polymerase. Thus the DNA sample may be repeatedly sequenced by allowing the primer extension reaction to continue for many cycles completely around the circular DNA sample. The data for this strand can then be converted into a more accurate consensus sequence with reduced data processing. In a distinct advantage over a system which employs detection of fluorophores, the system in this aspect uses the full capability of the read length of the polymerase, unhindered by having the read length reduced by phototoxicity.

A single molecule is a case of a monoclonal population in which the population is 1. As such ideas that are relevant to monoclonal DNA typically also apply to the single molecule condition situation.

Figure 25 describes and illustrates a device and method whereby a single DNA molecule 2507 can be sequenced by a NanoNeedle biosensor array 2500. A polymerase enzyme 2506 may be attached to a sensor 2501. A DNA sample with associated primers may then be caused to enter the volume with said polymerase attached sensors, utilizing for example, pressure induced flow, electrophoretic induced flow and or migration, or similar means. A single molecule from the DNA sample 2507 may then be bound by a polymerase attached to a sensor 2501 in a sensor array 2500. Additional single DNA molecules 2507 may also be bound by other polymerases 2506 bound to sensors 2501 in the sensor array 2500.

In one embodiment, one of the four native dNTPs 2502 is then flowed into the channel volume 2504 with the sensors. If the dNTP is complementary to the next base

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 in the sample DNA 2507, it may be bound and incorporated. The NanoNeedle sensor 2501 may then detect the resulting change in the local charge of the extended primer DNA, permitting detection of the incorporation event, at each appropriate position of the sensor array 2500. If the sample has more than one base in a row which is complementary to the type of dNTP 2502 which has been introduced into the channel volume 2504 with said sensors 2501, a second or subsequent binding and incorporation of a dNTP 2502 may be detected by said NanoNeedle sensors 2501. The dNTPs 2502 may then be washed out of the channel volume 2504 containing the sensors 2501.

In certain embodiments, one of the four native dNTPs is then flowed into the volume with the sensors. If the dNTP is complementary to the next base in the sample DNA, it is bound and incorporated. The NanoNeedle sensor may then detect the resulting change in the local charge, which may be as a result of the change in charge of the extended primer DNA, or may be as a result of other charge changes, permitting detection of the incorporation event, at each appropriate position of the sensor array. If the sample has more than one base in a row which is complementary to the type of dNTP which has been introduced into the volume with said sensors, a second or subsequent binding and incorporation of a dNTP may be detected by said NanoNeedle sensors. The dNTPs may then be washed out of the volume containing the sensors.

A different dNTP may then be flowed into the sensor array volume, permitting detection of incorporation events. Subsequent cycles of washing, introduction of each of the four dNTPs one at a time, and detection of incorporation events permit determination of the different sample DNA sequences.

In yet another embodiment, up to four different nucleotides may be delivered simultaneously, and determination as to which nucleotide is incorporated may be determined by observation of the kinetics associated with the incorporation reaction.

In an alternative embodiment, the sample DNA may be bound to one of the polymerase, the sensor, or a region between the sensors sufficiently close to the sensor that the bound polymerase may bind the sample DNA after a primer has been introduced into the system and permitted to hybridize with the sample DNA. Subsequently, after completion of the primer extension and associated determination of the sample DNA sequence, the extended primer may be melted off by changing the temperature or pH of the solution, or both the temperature and pH of the solution in which the sample DNA is solvated. The sample may then be re-sequenced by reintroducing the primer and restoring the temperature or pH of the solution in which the sample DNA is solvated to the conditions appropriate for primer extension, including appropriate concentrations of nucleotides and cations.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments, the nucleotides may be native dNTPs. In other embodiments, the dNTPs may be modified, with charge modifying structures. The charge modifying structures may be associated, bound or conjugated to the polyphosphate, and subsequently cleaved as part of the incorporation process, obviating the need for a separate process to cleave, separate, or remove the charge modifying structure.

In an alternative embodiment, the charge modifying structure is a terminator and thus be associated, bound or conjugated to the 3’ position of the sugar of the dNTP, and may thus act as a terminator. Detection may occur as a result of the process of incorporation, or may result from cleavage of the charge modifying structure.

In other embodiments, the charge modifying structure may be associated, bound or conjugated to the 2’ or 4’ positions of the dNTP sugar. In yet further embodiments, the charge modifying structure may be associated, bound or conjugated to the base of the nucleotide. The charge modifying structures may act as terminators, preventing the incorporation of additional dNTPs.

The linkage, association or conjugation may be broken as a result of a physical process, such as temperature change, or may be broken as a result of a chemical process, or may be as a result of a photochemical reaction. The linkage, association or conjugation may be broken after each nucleotide incorporation, or several nucleotides may be incorporated, and the number of nucleotides which were incorporated may be determined as a result of measuring the amount of charge which was added as a result of said incorporation(s).

In a further embodiment, two or three nucleotides at a time are used, allowing the addition of multiple bases at a time, and a correspondingly large signal. After completing the extension of the primer, with associated data collection, the extended primer is melted off, new primer added, and the process of extension may be performed again using a different order of combinations of dNTPs. This process determines which dNTPs do not follow the completion of a previous set of dNTPs, along with information as to the length of the incorporation, wherein said length determination need not be exact.

In order to permit repeated measurements of the same DNA sample, the DNA sample may be circularized, whilst the polymerase may be a strand displacing polymerase, or may be a polymerase with 5’ to 3’ exonuclease activity. Thus the DNA sample may be repeatedly sequenced by allowing the primer extension reaction to continue for multiple cycles around the circular DNA sample. In a distinct advantage over a system which uses detection of fluorophores, the system in certain embodiments of the current invention can employ the full capability of the read length of the

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 polymerase, unhindered by having the read length reduced by phototoxicity. In some embodiments, a strand displacing enzyme may be used, thus generating an increase in charge and associated counter ions. In other embodiments a polymerase with 5’ to 3’ exonuclease activity may be used, allowing net charge to remain the same, while generating protons and or hydroxide ions, which may be measured as an increase in conductivity, or may be measured as a result of the ions interaction with the surface of an ISFET, ChemFET, or NanoBridge sensor.

The polymerase bound or associated with the sensor may be a highly processive polymerase, permitting more bases to be incorporated then might occur with a less processive polymerase. The polymerase may be phi29, RcpliPHl, MagniPhi®, QualiPhi®, T4 (E. coli T4), F-530, Bl04, or other highly processive polymerases. The polymerase may be modified, so that it has reduced or no 3’ to 5’ exonuclease activity, or the polymerase may have no or little 3’ to 5’ exonuclease activity in its native form. Similarly, any 5’ to 3’ exonuclease activity may be modified so that it is reduced or virtually eliminated. Thermostable polymerases or other types of DNA or RNA polymerases may be used such as: Vent (Tli/Thermoccus Literalis), Vent exo-, Deep Vent, Deep Vent exo-, Taq (Thermus aquaticus), Hot Start Taq, Hot Start Ex Taq, Hot Start LA Taq, DreamTaq™, TopTaq, RedTaq, Taqurate, NovaTaq™, SuperTaq™, Stoffel Fragment, Discoverase™ dHPLC, 9⁹Nm, Phusion®, LongAmp Taq, LongAmp Hot Start Taq, OneTaq, Phusion® Hot Start Flex, Crimson Taq, Hemo KlenTaq, KlenTaq, Phire Hot Start II, DyNAzyme I, DyNAzyme ΙΙ,Μ-MulV Reverse Transcript, PyroPhage®, Tth (Thermos termophilus HB-8), Tfl, Amlitherm™, Bacillus DNA, DisplaceAce™, Pfu (Pyrococcus furiosus), Pfu Turbo®, Pfunds, ReproFast, PyroBest™ , VeraSeq, Mako, Manta, Pwo (pyrococcus, woesei), ExactRun, KOD (thermococcus kodakkaraensis), Pfx, ReproHot, Sac (Sulfolobus acidocaldarius), Sso (Sulfolobus solfataricus), Tru (Thermus ruber , Pfx50™ (Thermococcus zilligi), AccuPrime™ GC-Rich (Pyrolobus fumarius), Pyrococcus species GB-D, Tfi (Thermus filiformis),Tfi exo-, ThermalAce™, Tac (Thermoplasma acidophilum), (Mth (M. thermoautotrophicum ),Pab (Pyrococcus abyssi), Pho (Pyrococcus horikosihi, Bl03 (Picovirinae Bacteriophage Bl03), Bst (Bacillus stearothermophilus), Bst Large Fragment, Bst 2.0, Bst 2.0 WarmStart,Bsu, Therminator™,Therminator™ II, Therminator™ III, Therminator™ Y, T7 DNA,E. coli Polymerase I, Kenow (E. coli) Fragment, Klcnow fragment exo-, T4 DNA, Sulfolobus DNA Polymerase IV, AMV Reverse Transcriptase, human polymerase mu, human polymerase mu-h6, DNA Polymerase I (E. coli),T7 RNA (E. coli T7), SP6 (E. coli SP6)RNA, E. coli Poly (A), Poly (U), T3 RNA.

The polymerase and or DNA may be directly bound to or near the sensor, or may be bound through a linker.

In some embodiments, a variant of recombinase polymerase amplification as described in US7,270,981, is used for sequencing. In some embodiments, the amplified DNA template may be double stranded, and the input primers may be complexed with a recombinase such as RccA or RAD51. Said complexed primers may bind to the double stranded DNA, and with the aid of the recombinase, may displace a portion of the two strands of said double stranded DNA. A polymerase may then bind to the appropriate end of the primer such that said polymerase is able to incorporate nucleobases and extend said primer. Single stranded binding proteins may be added, which bind to the strand of the double stranded DNA which is not hybridized to the primer. As a result, a larger number of counter ions may be present for sensing, whereby, said counter ions may be associated with the newly synthesized strand of DNA, and additional counter ions may be associated with said single stranded binding proteins. An additional advantage is that issues due to secondary structure resulting from the use of single stranded DNA template are reduced.

Chamber-Free Reactors and Virtual Reactors

In other aspects, the invention provides a chamber-free device for sequencing a polynucleotide. The device comprises an electromagnetic sensor, a magnetic carrier for carrying or holding a template polynucleotide to or near the electromagnetic sensor, and a mechanism for removing the magnetic carrier via liquid flow and/or electromagnetic removal. In certain embodiments, the electromagnetic sensor is one of a nanoneedle or a nanobridge, and the device further comprises a local amplifier. The electromagnetic sensor may have a narrow structure, and maybe etched under the structure such that both sides of the sensor’s surface are accessible to changes in pH, or to changes in conductivity.

In some embodiments, the system employs magnetic arrays, as described in US Provisional Application 61/389,484 titled “Magnetic Arrays for Emulsion-Free Polynucleotide Amplification and Sequencing.

The system is shown diagrammatically in Figure 5.

Figure 5 depicts a single element 500 in an array of NanoNeedles and magnets 516, wherein a substrate 504 may have an electrode 506 on said substrate 504 and under a bead 502, or on a spacing or adhesion layer (not shown) between said electrode 506 and said substrate. Said electrode 502 may thus be within the Debye layer of the bead. A dielectric layer 508 may be placed above said substrate 504, and may also cover part of said electrode 506, and may further have a recess or cutout which may be larger than the space needed for said bead 502 when said bead 502 is retained in place. Said dielectric may serve multiple functions, including, providing a surface against which the

CA 2837486 2018-10-02 magnetic force may pull the surface of said bead 502. The magnetic force resulting from the interaction of magnets 516 and said bead 502, serve to retain and position said bead 502, with a force exerted downward against said electrode 502 and against dielectric 508. A second layer of dielectric 512 may be applied to said dielectric 508, providing significant reagent access to said bead 502, while further extending the height of the total thickness of dielectric material, such that an upper electrode 514 may be fabricated on said second layer of dielectric 512. Said upper electrode 514, dielectric, and second dielectric 512, may be positioned such that said upper electrode may be within the Debye length of said bead, and may further be close to the midpoint of said bead 502. Said upper electrode 514 may be above the center line of said bead 502, particularly if said upper electrode 514 is within the Debye length of said bead 502, or may be below the centerline of said bead 502, such that the top of said upper electrode 514 is in contact with said bead at an angle of between 10 and 90 degrees from the perpendicular of the point of contact between said bead 502 and said electrode 506. Said angle from the perpendicular to the point of contact between said bead 502and the top of said upper electrode may be between 30 and 85 degrees, between 45 and 80, or between 60 and 75 degrees. Said magnets 516 may be recessed into said substrate 504, be upon said substrate 504, or upon dielectric 508, but should be below the centerline of said bead 502, such that a downward force is applied to said bead 502, such that said bead 502 is pulled down towards said electrode 506. Said magnets 516 should further be placed offset with respect to the center of the bead 502 when said bead is in place in the array, such that said bead 502 is pulled towards said upper electrode such that said bead 502 is brought to within a Debye length of said bead 502 to said upper electrode 514. Each of the elements of single element 500 in an array of NanoNecdles and magnets 516 may have additional spacer layers or adhesion layers between said elements. Said Debye length may include Debye lengths which may result from high concentrations of salt, from low concentrations of salt, from deionized water, or from aqueous solutions which are comingled with nonaqueous fluids which are miscible in water.

Figure 2 is a photomicrograph of a combined virtual well and magnetic array according to various embodiments as described herein. Most positions in said array have a single bead in the position between the magnets at the point wherein the virtual well structure is located. Some locations have more than a single bead. Most of the ends of the magnets also have beads located thereupon.

The magnetic array may be used in a manner similar to that described in US 7,682,837.

As used herein, “bead” means beads, moieties or particles that are spherical or non-spherieal, where said beads, moieties or particles arc porous or solid or a mixture of

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 solid and porous, and can include magnetic beads that may be may be paramagnetic, super-paramagnetic, diamagnetic, or ferromagnetic.

As used herein, “bead capture features” means features that can temporarily hold a single bead in a fixed position relative to a sensor and can include local magnetic structures on the substrate, depressions which may include an external magnet, local magnetic structures, Van dcr Waals forces, or gravity as forces that fix the position of a bead. The bead may be bound in place by covalent or non-covalent binding.

As used herein, “confinement” refers to when a molecule generated (such as DNA) at one bead or particle stays associated with the same bead or particle so as to substantially maintain the clonal nature of the beads or particles.

As used herein “isolate” mean the prevention of migration, diffusion, flow, or other movement, from one virtual well to another virtual well as necessary to maintain the clonal nature of the beads or particles.

As used herein, “localized magnetic feature” means a magnetic feature created on a substantially planar substrate to hold individual beads on said substantially planar substrate.

As used herein, “localized magnetic field” means a magnetic field that substantially exists in the volume between the north pole of a first magnetic region and the south pole of a second magnetic region or substantially exists in the volume between the north and south poles of a single magnetic region.

As used herein, “particle” means a non-bead moiety such as a molecule, an aggregation of molecules, molecules bound to a solid particle, or particles, and other forms known in the art.

As used herein, “single phase liquid” is a liquid with relatively uniform physical properties throughout, including such properties as density, index of refraction, specific gravity, and can include aqueous, miscible aqueous and organic mixtures but does not include non miscible liquids such as oil and water. Among the physical properties not considered to potentially cause a liquid to not be considered a single phase liquid includes local variations in pH, charge density, and ionic concentration or temperature.

As used herein, “substantially planar” shall allow small pedestals, raised sections, holes, depressions, or asperity which does not exceed 40 gm relative to the local plane of the device. Variations due to warpage, twist, cupping or other planar distortions arc generally not considered to constitute a portion of the permitted offset. Protrusions or depressions which may be not essential for the uses as described herein but which exceed 40gm do not preclude a device from being considered substantially planar. Fluidic channels and or structures to generate said fluidic channels which have

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 dimensions of greater than 40gm also do not preclude a device from being considered substantially planar.

As used herein, “virtual wells” refer to local electric field or local magnetic field confinement zones where the species or set of species of interest, typically DNA or beads, generally does not migrate into neighboring “virtual wells” during a period of time necessary for a desired reaction or interaction.

As used herein, “electrode” is defined as any structure used for creating or applying the electric or magnetic force in such array. Such a structure may be used for isolation or manipulation in the delivery of a biomolecule to a special region in the array (e.g. the middle region of an element in the confinement or isolation array) at a time of interest resulting from turning on and off of the fields or forces.

In embodiments of the devices disclosed herein, the device comprises a sensing surface for sensing incorporation of a nucleotide, the sensing surface comprising a layer of silicon nitride.

In embodiments of the devices disclosed herein, a plurality of magnetic beads arc configured for carrying template polynucleotide, wherein the magnetic beads have a low zeta potential material at a pH level effective for nucleotide incorporation.

The a virtual nanorcactor or “chambcr-frcc array,” may detect or manipulate particles (e.g., beads, cells, DNA, RNA, proteins, ligands, biomolecules, other particulate moieties, or combinations thereof) in an array wherein said array captures, holds, confines, isolates or moves the particles through an electrical, magnetic or electromagnetic force and may be used for a reaction and or detection of the particles and or a reaction involving said particles. Said “virtual nanoreactor” provides a powerful tool for capturing/holding/manipulating of beads, cells, other biomolecules, or their carriers and may subsequently concentrate, confine, or isolate moieties in different pixels or regions of the array from other pixels or regions in said array using electrical, magnetic, or electromagnetic force(s). In one embodiment the array is in a fluidic environment. Sensing may be done by measurement of charge, pH, current, voltage, heat, optical or other methods.

The chamber-free device described herein in certain embodiments allows for better washing of nucleotides during sequencing reactions, may reduce leading sequencing phase error, by reducing the number of residual nucleotides which may include both unbound and nonspecifically bound nucleotides, which may later be inappropriately incorporated in an incorrect cycle.

In addition to DNA sequencing, various different molecular biology applications using the ’’chamber free” array or “virtual” nanoreactor are envisioned. The array may be used as a tool, for example, as a cell sorter, and subsequently the array may 33

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 additionally perform molecular biology on said cells, which may include sorting, measurement or manipulation for one or more biochemical events or reactions of interest (e.g. drug screening, or biomolecular detection). The virtual nanoreactor array may be used for cell monitoring and analysis, for example, the system may measure the electrical signature of a cell where said cell is captured in the array and or adjacent to sensing elements associated with the virtual nanoreactor. The array may be used for screening of rare cells, for example, for detection of reactions in drug screening for drug development, or for selection and testing of cancerous cells. In various embodiments assays or detection targets include cell biology, drug screening and monitoring for specific cell types, detection of DNA, RNA (nucleic acids), proteins, charged small molecules, ligands, or other biomolecules.

In other embodiments, a further electrode (or virtual wall/fence) element may enclose the other two electrodes associated with each electrical confinement or isolation element in the array.

In some embodiments, the system is used to capture multiple beads or cells per pixel, and may turn on, tum off or modify the magnitude, shape or period of the electromagnetic field at a desired time or in response to a change at a sensor associated with said element in the array of virtual walls as may be needed for different applications. For example the system may capture one set of beads or cells and then increase the field strength to capture another moiety which is less influenced by said field. The electric field can be a DC or AC or combination of different combinations of the fields.

Figure 3 is a graphical representation resulting from a Comsol simulation, depicting the 3D equipotential field strength curves 306, of one fourth of a cylindrical structure 300 that result from a field being applied to the electrodes 304A and 304B. As a result of the substantial radial asymmetry of the field, the volume where the field gradient is most concentrated 308 near the center electrode 304 B, which is the point where a bead may be held by a magnetic array as shown in Figure 2.

In other embodiments, the virtual nanoreactor(s) are used for and or combined with multiple steps in a biomolecular process, for example, bead enrichment of particle(s) moving in an electric field, micro fluidic sample-prep and library prep such as on-chip extraction of DNA, shearing of DNA, on-chip normalization of DNA concentration, emulsion-free amplification, sensing, which may include SensePlus sensing utilizing dual-sensing or multi-sensing sequencing detectors, which may utilize transient and or steady state electronic sequencing and rephasing methods for extending the length of read of a sequencing process, wherein said multiple steps may result in a fully integrated electronic genomic analyzer system.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments, multiple arrays of virtual nanoreactors are used. The different arrays of virtual nanoreactors may be used for different biological reactions, processes or methods. In some embodiments, one array or set of arrays is used for extraction of DNA, where different arrays in said set of arrays or different members within a set may retain cells from different samples, or may retain different types of cells from a single sample, or a combination thereof. For example, different samples may be held in different arrays, and different types of cancer celts from a single sample tumor may be retained within different areas of a single array. The different cell types may be sorted or segregated, or the cell type may be determined as a result of data derived from the individual cells after said cells are captured and retained in individual locations on said array of virtual nanoreactor.

In some embodiments, several biological reactions, processes or methods may be performed on a single sample or moiety while it is retained in position within a virtual nanoreactor. In other embodiments, one or more biological reactions, processes or methods may be performed on a samplc(s) or moicty(s) in a positions) in a virtual nanoreactor array(s) prior to the transfer or movement of said sample(s) or moiety(s) to another position(s) in another virtual nanoreactor array(s).

In some embodiments, the array of electrical concentration and confinement virtual nanoreactors is used for purposes other than the capture/isolation of nucleic acids as otherwise described herein for sequencing and amplification of nucleic acids. In some embodiments, the electrical concentration and confinement is used with any charged moiety, where the time period, moiety concentration, mobility, concentration, local viscosity, cross-linking percentage of local polymers, or concentration of the charged moiety(s) influences the spacing of electrical confinement structures, and or field strength used, and or magnitude of field gradient in order to maintain sufficient confinement within individual virtual nanoreactors. The level of cross contamination between different virtual nanoreactors which may be tolerated may vary for different applications, and at different steps in a biological reaction, process or method. For example, if a nucleic acid amplification reaction process was performed on different samples in adjacent virtual nanoreactors, cross contamination may be problematic during early thermal cycles of a PCR reaction, while cross contamination during later cycles of a PCR reaction may not be particular concern, as primers may be largely consumed, preventing significant amplification of the cross contaminating nucleotides from other virtual nanoreactors.

In some embodiments, a polymer is used to reduce the migration rate of nucleotides or other moieties. For example, the polymer may be a partially entangled polymer such as POP-7™, or an agarose, polyacrylamide, or starch gel. The concentration and or percentage of cross linking may be varied as needed for various

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 mobilities of different moieties for which confinement in the virtual nanoreactor is desired. The polymer may be introduced to the volume of the virtual nanoreactor before, with, or after the introduction of sample moieties or other moieties used in said biological reaction, process, or method.

In some embodiments, the electrical concentration and confinement is used to capture moieties to which other biomolcculcs or biological moieties have been bound, for example beads with charge to which antibodies may be bound, and wherein said beads may be subsequently used to capture proteins, and wherein said beads further may be subsequently captured using the electrical concentration and confinement array.

In some embodiments, the virtual nanoreactors may be used for several different applications in addition to sequencing, for example said virtual nanoreactors may be utilized as a hybridization array (similar to an Affy or Agilent DNA micro array) wherein the DNA is on beads, and said beads are held in place by the virtual nanoreactor. In other embodiments, the virtual nonreactors are used for digital PCR, where said beads are introduced into the array. Detection may be done with an electronic sensor array associated with each element of the array, or detection may employ optical means, to detect the presence or quantity of a particular nucleic acid. Digital PCR may be used to quantify the concentration of targets within a sample in a relativistic manner.

In some embodiments, concentration of sample into the virtual nanoreactors as described herein allows quantitation with greater sensitivity. In some embodiments, using concentration in some areas of the array and not in others may allow for a greater concentration quantitation dynamic range. In further embodiments, DC concentration fields may be inverted so as to partly “push” sample moieties away from said virtual nanoreactors. In some embodiments, combinations of concentration, no concentration, and pushing away of sample moieties to different areas may help to further extend the dynamic range of digital PCR, or of any other desired biological reaction, process, or method. Beads with different primers may be introduced into different areas of the virtual nanoreactor array as described herein, allowing simultaneous digital PCR reactions for several targets.

In some embodiments, the virtual nanowells are used for full extension reactions of DNA bound to primers conjugated to said bead. In a further embodiment, said virtual nanoreactor may be used for ligation reaction detection.

In some embodiments, said system may use electromagnets, permanent magnets, or electrodes or other different subsystems to generate electromagnetic fields for transient or part-time isolation, or holding or concentration of the biomolcculcs or other moieties of interest (e.g. DNA, cells, proteins) or the carrier of the biomolecule or other

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 moiety of interest (e.g. beads, particles, or other moieties). Said electromagnetic fields may be magnetic fields, electric fields, or a combination of the two. Said electric fields may be DC fields, AC fields, pulsed DC fields, non-sinusoidal AC fields, pulsed AC fields, or a combination thereof. The nanorcactors with virtual walls (or fences) may be used or created with electric or magnetic fields to hold, capture, concentrate, isolate, or manipulate the biomolecules or its carrier.

In some embodiments, where two or more sets of electrode structures are employed, virtual walls or isolation walls or fences may be turned on or off, or modified as to the magnitude, shape or period of the electromagnetic field at a desired time, which may be a fixed time, or may be in response to a change from a sensor associated with a specific member of the array of electrode structures. Turning on or off, or modifying as to the magnitude, shape or period of the electromagnetic field may be used for controlling the movement of the particles, beads, cells, biomolecules or other moieties of interest which are concentrated, confined, or isolated in the array of electrode structures, or for controlling the biomolcculcs or other moieties of interest which may be used in a reaction, such as an antigen or secondary antibody in protein detection, or nucleotides in DNA or RNA sequencing, or secondary cell in cellular interactions, or drugs or cells in drug screening and monitoring. This feature may be used to provide easy access and flexible manipulation, and or mixing.

In some embodiments, the virtual nanoreactor array normalizes the amount and/or concentration of DNA input into the system, and may generate a feedback and/or control to the entry, which may then be used to control the amount and/or the concentration of said DNA, or other biomolecules or other moieties input into the system as described herein. The system may be further used for the purpose of realtime normalization of the detection or sequencing array.

The array for capturing (isolating, confining or concentrating) of beads, or cells, or other biomolecules, particles, or other moieties of interest through magnetic or electromagnetic or electric capturing and or holding may be structured such that it comprises two sets of the capturing elements (GENIUS bars or elements of a magnetic array as described elsewhere) per element of the array, allowing the capture (isolating, confining or concentrating) of one moiety with one set of capturing elements, and the subsequent capture (isolating, confining or concentrating) of a second moiety of interest. Said capturing can be done in different orders and with different structures, and may include more than two sets of capturing elements per array element, and correspondingly, additional capture steps may be performed at each array element.

In some embodiments, the electrical concentration and confinement array may be used to capture charged beads to which B-cells have been bound, for example magnetic beads to which proteins, polysaccharides, or other immunogens are carried,

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 and where the beads may be subsequently used to capture B-cells. The charged beads may be subsequently captured using said electrical concentration and confinement array. In some embodiments, the electrical concentration and confinement array is used to capture B-cclls to which proteins, polysaccharides, or other immunogens may be bound.

In some embodiments, the electrical concentration and confinement array is used to capture charged beads to which other carbohydrates or glycolipids have been bound, for example charged beads to which proteins or peptides which comprise carbohydrate binding modules may be bound. The charged beads may be subsequently used to capture carbohydrates or glycolipids, and where the charged beads further may be subsequently captured using said electrical concentration and confinement array. Such a carbohydrate or glycolipid binding moiety may comprise a carbohydrate active enzyme, such as a glycosidic hydrolase, a lectin, a galectin, a intelectin, a pentraxin, a selectin, an adhesion, or a hyaluronan.

In some embodiments, the array is used for chemical screening applications and non-biomolecules can be tested or monitored or measured with the system. For example, in use for drug screening, where measurement of a drug effect is desired, the system may employ, for example, about 100,1000,10,000, or 1,000,000 different drug candidates each on their own beads, where said beads may be captured and hold in the virtual nanoreactor array. The system may subsequently allow' interactions and measurements of the interactions between the beads and a cell or set of cells of interest, where the electrical confinement is used for isolation of the pixels in the array, providing a high throughput and fast drug screening system.

In some embodiments, the electrical concentration and confinement array is used to capture a combined set of charged beads to which multiple different types of biomolecules have been bound. The combined set of charged beads may comprise multiple sets of charged beads w'ith different types of biomolecules or biological moieties bound. The may be one type of binding moiety which may bind one biomolecule or biological moiety on a set of charged beads, and a different type of binding moiety which may bind a different biomolecules or biological moiety on a different set of beads. Said combined set of charged beads may also comprise sets of charged beads w'hcrcin the set of charged beads may comprise charged beads to which multiple binding moieties are bound.

In further embodiments, said combined sets of beads may further comprise a label, wherein the label differentiates between different sets of beads. Said label may be an optical label, such as a fluorescent dye, a biochemical label such as DNA, metal particle labels, any other type of label, or a combination of different types of labels.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Said labels may be used to determine which type of bead may be at each element of the array electrical concentration and confinement features.

In some embodiments, different methods of detection of the beads and the interactions that result from various biomolecular reactions may be observed as a result of the detection of said beads. Said detection may be effectualized as a result of a NanoNccdlc sensor, a NanoBridge sensor, a ChcmFET sensor, an ISFET detector, an optical sensor, such as for example a fluorescence detector, a SERS detector, an absorption detector, a PH detector, a conductance detector, a mass resonance detector, a calorimeter detector, or any other type of detector suitable for detection of biomolecular reactions, or of other types of reactions. In some embodiments, said sensors may be combined at each position in the array with an electrical concentration and confinement feature.

In some embodiments, the electrical concentration and confinement are used to directly capture charged moieties, where said charged moieties are bound to other biomolecules or biological moieties. For example antibodies or antibodies bound to proteins may be concentrated or confined ,where said antibodies or antibodies bound to proteins may be subsequently used to capture proteins or additional antibodies.

In some embodiments, the virtual nanoreactor array increases the reaction rate by concentrating biomolcculcs of interest, such as DNA, RNA, or other reactants or reagents for more efficient synthesis.

In some embodiments, it is desirable to integrate a valving system as part of a flow cell. Said valving system enables the flow of samples to sections of a flow cell, such that different samples may be used for different sections of said flow cell. In other embodiments, the valving system is integrated adjacent to the flow cell, whereby the valving system and flow cell may form a scaling interface to each other. In other embodiments, said valving system and said flow cell may be adjacent to each other on a single mount, wherein both said valving system and said flow cell may be mounted to said mount. Said valving system may also comprise a waste valve(s) such that fluids may be removed from said valving system prior to flowing into said sections of said flow cell. For example if there is a significant dead volume in said valving system, it may be desirable to remove fluid which may have an unacceptable level of cross contamination from a previous fluid.

In some embodiments, it may be desirable to integrate a valving device with the flow cell. Such a valving configuration may include various inputs, which may include inputs for the four dNTPs (e.g., for sequencing reactions), which may also contain buffer, salts, enzyme and any other moieties required for incorporation of nucleotides. Inputs may also be employed for various buffers and wash reagents, polymerase

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 containing buffers, which may also contain salts and any other moieties needed for polymerization, reagents needed to strip any coatings from the flow cell, reagents which may be needed to re-coat the flow cell, buffers which also include a phosphatase, or other reagents.

In one embodiment, the valving device is fabricated of PDMS. In another embodiment the valving device is fabricated from glass with magnetically or pneumatically activated elastomeric valves.

In some embodiments, it may be desirable to bond said valving and fluidics PDMS manifold to a silicon device. It may be desirable to increase the bonding strength between said PDMS and said silicon device.

In one embodiment of the current invention, it may be desirable to use plasma activated PDMS to improve the bond strength. As a plasma treatment which has too much power or too much pressure may actually decrease the bond strength of PDMS to silicon, it may be important to use lower power levels and pressures. Tang ct al describe appropriate power and pressure levels in 2006 J. Phys.: Conf. Ser. 34 155. In one embodiment it is suitable to use a pressure between 500 mili Torr and 30 miliTorr and a power level between 10 and 60 watts while using, for example, a 790 series Plasma-Therm.

For a device fabricated of PDMS or other similar materials, it is possible to use pressure valves to control the flow of reagents. With such valves it is possible to have the several valves in close proximity to each other, and the valves may be very close to a central channel, reducing dead volume, as shown in Figure 4A, which shows a reagent valve system 400 with three reagent input lines 402 with valves 406, each of which can be configured to flow' towards the input to the flow cell 408, under the control of pressure control lines 404.

For a more complex system, where more reagent inputs are desired, the simple valve system 400 of Figure 4A is insufficient, as it has but three reagent inputs lines 402. In an alternative embodiment as shown in Figure 4B and 4C, many more inputs are enabled. This approach also permits clearing the dead volume of the channel. In Figure 4B, inputs include input ports for dATP, dTTP, dCTP, dGTP, Buffer one, Buffer two, and Sample, output ports Waste one, Waste two, and Waste three. Control lines are in place for each input and output port, with additional control lines to control the direction of flow between activated ports. A waste port is shown immediately prior to the flow cell, so that any remnant reagent from a previous flow may be removed, allowing a very clean transition from one reagent to another, without diffusion from any dead volumes in the valving system. Figure 4C depicts a valving system with a oval flow path, such that all input valve port positions have a path to an outlet (waste) port in

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 both directions from said input valve port position. Valves as shown in Figure 4A may be used for each valve shown in Figure 4B, 4C or in a physical embodiment of a reagent valving system 430 as shown in Figure 4D, wherein photograph of a PDMS valve system 440 is shown.

One embodiment permits purging of air or other contaminants in each input line up to each input port control valve, so that when each input port is activated, appropriate reagents may be introduced to the system. For example, to clear the dATP line, the dATP control line and the Waste one Upper Control line may be activated causing air and any unwanted reagents in the dATP lines to flow through the dATP valve and out Waste one. In other embodiments, the dCTP, dTTP, and the dGTP lines may be purged or cleared of contaminants by activating the dCTP control line and the Waste one Upper Control lines, the dTTP and the Waste one Upper Control lines, and the dGTP control lines and the Waste one Lower Control line respectively. To purge or remove contaminants from the sample line, the Sample Control line and the Waste two Control lines may be activated.

Purging and removing contaminants from all lines may be needed after replacing a sequencing assembly. Similarly purging or removing contaminants from reagent lines may be needed after replacing or refilling reagent bottles or containers. A further need for purging or removing of contaminants may result from periods of time wherein the instrument is not used, which may permit any reagent lines containing reagents which need to be cooled below ambient to suffer degradation; for example, polymerase in a polymerase containing reagent may suffer from extended exposure to ambient temperatures.

In some embodiments, it is desirable to fill the manifold leading to the input of the flow cell, so that any reagents remaining from a previous use of the manifold may be removed. For example, prior to introduction of dATP into the flow cell, the dATP control line, the Upper Liquid Control line, and the Waste two Control line may be activated. dATP reagent will then commence to flow from the dATP input line, around both sides of the upper liquid loop, through the channel between the upper liquid region and the lower liquid region, and out through the waste two valve into the waste two line. Alternatively, prior to introduction of dCTP into the flow cell, the dCTP control line, the Upper Liquid Control line, and the Waste two Control line may be activated. dCTP reagent will then commence to flow from the dCTP input line, around both sides of the upper liquid loop, through the channel between the upper liquid region and the lower liquid region, and out through the waste two valve into the waste two line. Similarly, prior to introduction of dTTP into the flow cell, the dTTP control line, the Lower Liquid Control line, and the Waste two Control line may be activated. dTTP reagent will then commence to flow from the dTTP input line, around both sides of the lower liquid loop,

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 through the channel between the lower liquid region and the lower liquid region, and out through the waste two valve into the waste two line. Similarly, prior to introduction of dGTP into the flow cell, the dGTP control line, the Lower Liquid Control line, and the Waste two Control line may be activated. dGTP reagent will then commence to flow from the dGTP input line, around both sides of the lower liquid loop, through the channel between the lower liquid region and the lower liquid region, and out through the w'aste two valve into the waste two line.

Buffer one can be made to flow through the main flow cell (in dark blue), and out the Waste three port by activating the BIC control line, and the W3C control line. Alternatively, Buffer one can be made to flow out the Waste two port by activating the BIC control line, and the W2C control line. In another alternative use, Buffer one can be made to flow out the Waste one line through the upper section of the liquid manifold by activating the Buffer one control line, the Upper Liquid Control line, and the Waste one Upper Control line. Similarly, the lower liquid manifold can be flushed with Buffer one by activating the Buffer one control line, the Lower Liquid Control line, and the Waste one Lower Control line. Activating flow in a combination of these areas either in a time sequence or activated together enables clearing the entire liquid manifold to be purged of bubbles, other contaminants, or as a wash or purge of any other liquids which may have been introduced to the system via the dATP, dTTP, dCTP, dGTP, Buffer two, or Sample input ports.

In some embodiments, it is desirable to use a passivation layer over the silicon device which has a higher bond strength than thermally grown silicon dioxide. Tang et al describe several passivation layers which provide improved bonding strength, including PSG (PECVD phosphosilicate glass), USG (PECVD undoped silicate glass), ShN^LPCVD silicon nitride).

In some systems which use PDMS valving manifolds, pins or needles inserted into the PDMS are used to connect reagent lines to the valving manifold. While this provides for secure attachment, attaching a number of reagent lines to a PDMS valving manifold is time consuming and error prone. Thus in some embodiments, it may be desirable to use an interface manifold, where the reagent lines are connected to the interface manifold, rather than to the valving manifold, and the interface manifold may be connected to the valving manifold. The reagent lines may be attached to pins or needles, which may be attached to the interface manifold. The pins or needles may be permanently affixed to the interface manifold, being held in place with an adhesive, by welding or brazing, by utilization of a press fit, or by some other means. Alternatively, the lines may be directly connected to the interface manifold, where they may be retained by a fitting or o-ring, or by some other means as known in the art.

In some embodiments, the interface manifold may scalingly interface to the valving manifold such that reagents may flow from the interface block to the valving manifold. The interface between the interface manifold and the valving manifold may be an interface which is used by a user to enable replacement of the chip/flow cell and/or valving manifold.

In some embodiments the interface manifold may have internal channels formed by bonding, such bonding could include fusion bonding, solvent bonding or adhesive bonding.

In some embodiments, minimizing the path length to the active part of the flow cell may be important for several reasons, including minimizing the amount of mixing of reagents, which occurs as a result of differences in flow rates at the center of a channel versus the flow rate at the edges of a channel, due to wall interactions, as well as diffusion. In some embodiments it may also be desirable to minimize the volume which is not temperature controlled, in order to prevent degradation of reagents, such as polymerase, in volumes which are not temperature controlled. In some embodiments it may be desirable to minimize plumbing volume to concomitantly minimize cross contamination of reagents may also occur in regions of flow which are common to multiple reagents, due to nonspecific binding to materials which contact said reagents.

Figure 1 illustrates a schematic representation of one embodiment of the cunent invention, where a magnetic or paramagnetic bead is held in place over a sensing region by a magnetic array. The magnetic array is described in US Provisional Application 61/389,484 titled “Magnetic Arrays for Emulsion-Free Polynucleotide Amplification and Sequencing . Retained magnetic or paramagnetic beads may have monoclonal populations of DNA. The beads may be sized such that there is sufficient room for only one bead over each sensor, thus providing for a one to one correspondence between sensors and beads. Although there may be room for only one bead over each sensor, there can be room between beads when the beads arc aligned over the beads, resulting in reduced cross-talk between sensors. For example, a set of beads may be about 10 microns in diameter located over sensors which are about 8 microns across, and the sensors may be spaced about 15 microns apart, resulting in an approximately 5 micron space between the beads. The size of the sensors may be larger than the beads, if there is insufficient room for two beads to be retained above the sensor. The size of the beads, sensors, and spacing can vary. In other embodiments, beads may be greater in size than 10 microns, such as about 15 microns, about 20 microns, about 25 microns, or larger. In further embodiments the beads may be smaller than 10 microns, such as about 5 microns, about 3 microns, about 2 microns, about 1 micron, or less than one micron. The sensors may be sized to align with the size of the beads, and thus may be larger, or smaller in size

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 than 8 microns across, potentially ranging from less than one micron, to about 1, 2,3, 5, 10, 15, 20 or more microns across. The spacing between the sensors may also be greater than 15 microns, or may be less than 15 microns; the sensor spacing may range from less than one micron, to about 1, 2, 3, 5, 10,15, 20, 25 or more microns between sensors.

A chamber-free magnetic retention structure as shown in Figure 1 may permit improved flow of nucleotides, polymerase and other components, as their flow is not hindered by a well structure, such as that shown in Figure 2, permitting better washing, more complete incorporation of bases, and faster cycle times then w'ould be possible if the bead were in a well. In a well structure, a bead and associated DNA such as shown in Figure 2 hinders accessibility and flow, so that a higher concentration of polymerase and nucleotides may be needed to permit sufficient diffusion to all parts of a bead as shown in Figure 2. Said higher concentrations of dNTPs and polymerase may increase the error rate due to misincorporation by said polymerase, resulting in higher levels of leading sequencing phase error then might occur with a chambcr-frcc structure such as is shown in Figure 1.

In some embodiments, the array described herein is reusable (e.g., not single use). The cost of sequencing has a number of parts; for sequencing using electronic sensors, one of the major costs is the cost of the processed silicon itself; that is: the sensor. This may be particularly true if the sensor is not re-useable, but must be discarded after a single use. The magnetic array described above makes reuse fairly straightforward, as the DNA is not bound to the sensor, and the beads can be easily removed by reducing or removing the magnetic field which holds said beads in place. If the beads are instead held in place with a structure, removal may be more difficult.

In one embodiment, beads which are held in place by a structure, or array of wells, and removed by applying a magnetic field such that the beads, which may be magnetic or paramagnetic beads, are pulled out of the wells, and subsequently removed from the flow cell by flowing a reagent through the cell.

In some embodiments, the array of magnetic features are used for purposes other than the capture/isolation of nucleic acids as otherwise described herein for sequencing and amplification of said nucleic acids. These include capture of cells (e.g., cancer cells, B-cells), proteins, glycoproteins, glycolipids, antibodies, saccharide or polysaccharide, and other moieties as already described.

In some embodiments, associated cells bound to retained beads are lysed while the beads are held in position in the magnetic array. The retained beads may further comprise attached primers or primer sets for amplification and/or sequencing target nucleotide sequences. The primers may be universal primers, or primers targeted to

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 specific sequences associated with the type of cell bound to each bead, or may be universal primers or universal primer sets with barcodes. For example, barcodes are associated with the cell type bound to each bead, or a combination of different primer types. In some embodiments, after lysing of cells, a reverse transcription and/or amplification reaction may be performed. In some embodiments, the amplification is a real time PCR reaction, where the quantity of a specific RNA may be determined for each bead and thus each cell type. In other embodiments, the amplification reaction is a PCR reaction or isothermal reaction, and may generate clonal populations on said beads, or may generate multiclonal populations, where each clone type may use different primers. In other embodiments, a sequencing by synthesis reaction is subsequently performed to determine the sequence of the amplified sequence(s) associated with each bead type, and thus with each cell type.

In further embodiments, after lysis of cells, DNA, RNA or other molecules of a specific charge may be retained, and the beads may be removed. In some embodiments, additional beads may be introduced to said magnetic array. The newly introduced beads may have different primer types associated with the newly introduced beads as described herein.

In some embodiments, the array of magnetic features may be configured such that a preferred position is maintained by a magnetic or paramagnetic bead or particle. Such a preferred position may be desired so as to appropriately position the particle with respect to a sensor or sensors, and/or to maintain the particle in a fixed location, so that the particle and the charge attached to the particle does not move with respect to said sensor or sensors. In some embodiments the preferred location may result, at least in part from the configuration of the magnetic array element shapes, as shown Figure 6. Figure 6 illustrates two different configurations 600 wherein at least some of the shapes in the array may be configured such that the density of the magnetic flux is more concentrated at one end of members of the magnetic array than at the other end or some other members of the magnetic array. In the top two pairs of magnetic elements, the left trapezoidal magnetic array element 604 is narrower at the south pole, than at the north pole of said trapezoidal magnetic array element. As the total flux level emanating from the magnetic array element must be the same at the two ends, the flux density at the narrower end of the trapezoidal magnetic array element 604 will be higher than at the wider end of said trapezoidal magnetic array element. As a higher concentration of flux corresponds to a higher force exerted on a magnetic or paramagnetic element, a higher force will be exerted on the magnetic or paramagnet particles or beads 602 by the narrower end of said trapezoidal magnetic array element 604, than is exerted by either the similarly sized magnetic array element 606 on the right side wherein the element on the magnetic array element on the right side 606 is of similar size, but is rectangular, and thus has a lower flux density and force. Similarly, a higher force will be exerted on 45

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 the magnetic or paramagnet particles or beads 602 by the narrower end of the trapezoidal magnetic array element 604, than is exerted by either the similarly sized trapezoidal magnetic array element 608 on the right side wherein the element on the magnetic array element on the right side 608 has its wider end oriented towards the magnetic or paramagnetic particle or bead 602.

In another embodiment of the current invention, additional features may be incorporated as part of fabrication of the magnetic array so as to inhibit motion of the magnetic or paramagnetic particle, or bead 702 as shown in Figure 7. In the embodiment shown, the magnetic or paramagnetic particle or bead 702 is pulled preferentially towards the narrow end of the left trapezoidal magnetic array clement 704, and is pulled into contact with two posts 710, as well as being pulled down into contact with the surface of the array, thus providing three points of contact to fully stabilize said magnetic or paramagnetic particle or bead 702. The posts and magnetic array elements may provide minimal areas of surface contact, so as to permit maximal access by ions, dNTPs enzymes and other moieties. The posts and magnetic array members may be positioned with tight tolerances with respect to the sensor elements, so as to provide reproducible signal levels between different members of the sensor array.

In further embodiments, a small well is used, such that a magnetic or paramagnetic particle may rest on the upper comers of the well. The well may be round, or may be of some shape other than round if the magnetic or paramagnetic structure is generally spherical in shape, so as to allow better access to the bottom of the magnetic or paramagnetic particle by enzymes, dNTPs, ions and other moieties.

In some embodiments, the magnetic array is used to generate clonal populations for hybridization detection, hybridization pullout, or sequencing. The assay may be performed with beads in the positions in the magnetic array where the amplification occurred, or the beads may be moved from the area or volume where the amplification reaction took place to another location. The second location may also employ a magnetic immobilization to perform the assay, or may employ a different immobilization such as biotin streptavidin binding. In some embodiments, the sensors are positioned directly under the magnetic or paramagnetic particle, so as to maximize the interaction between the charge associated with the DNA on the bead and the sensor. In other embodiments, the bead may be bound or associated in such a manner that it is fixed and unable to rotate freely. It may be desirable to position the sensor off center from the paramagnetic, permitting access to areas of the particle which have free access to the aqueous environment and access to polymerase, dNTPs and other moieties is permitted and may have optimal enzymatic reaction which may then be read by the sensor, in contrast to the areas which are in direct contact with the surface where the reaction may be inhibited as a result of lack of access to the aqueous environment.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments, it is desirable to use the differential flow which occurs in a channel to rotate the magnetic or paramagnetic particles, so as to provide optimal access by enzymes, dNTPs and ions to all surfaces of said magnetic or paramagnetic particles. Said differential flow results from the parabolic flow typical of small channels, where the bulk flow rate at the surface of a channel is zero, and the maximal flow rate in the channel is typically highest in the center of said channel. This may result in a significant flow rate differential between the bottom of the channel and the top of the magnetic or paramagnetic particle. The difference in flow rate between the top and bottom of the magnetic or paramagnetic particle will be a function of the size of the particle, the height and width of the channel, and the average flow rate in the channel. The DNA and or other moieties which may be attached to the magnetic or paramagnetic particle may provide a drag or pull on the top of the magnetic or paramagnetic particle due to the comparatively high flow rate at the top of said magnetic or paramagnetic particle. The flow rate at the bottom conversely will remain essentially zero, thus creating a significant rotational impetus. The magnetic array elements, potentially combined with other physical features may maintain the position of said magnetic or paramagnetic particle.

In some embodiments, the flow rate is maintained at a consistent flow rate while introducing and flowing dNTPs and or other reagents while reading the sensors, thus maintaining a consistent average rotation rate. In other embodiments, it is desirable to reduce the flow rate while reading the sensor, so as to prevent significant oscillations and motions of the magnetic or paramagnetic particle. In yet other embodiments, it is desirable to increase the flow rate while flowing reagents through the flow cell when reading the sensor, so as to permit more surface area of the magnetic or paramagnetic particle to interact with the sensor. This may permit an increased averaging effect, which may reduce sensor readout variations due to variations in DNA attachment density on the surface of the magnetic or paramagnetic particle, or variations in the sensor readout due to irregularities in the shape of the particle.

In other embodiments, forces other than that resulting from changes in the flow rate velocity are used to modify the rate of rotation or movement of the magnetic or paramagnetic particle. In some embodiments, the flux levels may be modified as a result of the movement of an external magnet which may be coupled through the magnetic array elements as a result of the higher permeability of the magnetic array elements. Tn an alternative embodiment, and electromagnet is used to influence the amount of flux which interacts with the particles. These changes in the amount of magnetic flux may reduce the frictional forces acting on the magnetic or paramagnetic particle, permitting more or less rotation.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In yet other embodiments, a magnetic or electric field is used to rotate the magnetic or paramagnetic particles, in part as a result of the polarizability of the bead and associated DNA.

In some embodiments, it is desirable to use a force in addition to that which results from the flow of reagents through the flow cell to position magnetic or paramagnetic particles into appropriate locations associated in a one to one correspondence with sensors. In some embodiments, magnetic or paramagnetic particles may be flowed in a reagent stream into a flow cell or cells, and a magnet or electromagnet is used to move the particles into positions associated with sensors such that a higher proportion of the magnetic or paramagnetic particles is associated in a one to one correspondence with sensors than would have occurred without the use of such additional magnets or electromagnets.

The magnetic array also permits virtually complete allocation of beads to array locations. Low speed flow is sufficient to enable localized retention of the beads in the array, in a one to one correspondence between beads and array locations, without requiring centrifugation. In one embodiment, if even higher levels of filled versus unfilled locations on the array is needed or desired, the reagent flow may be circularized, such that the beads may be reintroduced to the flow cell. In another embodiment, the reagent flow may be stopped or slowed as the beads may be introduced to the flow cell. In another embodiment, the direction of the reagent flow may be reversed, potentially several times, providing more opportunities for the beads to fill the array. In yet another embodiment, the beads are retained after flowing the beads into a flow cell by flowing the beads through either the inlet or outlet to a storage location, so that they can be used in a subsequent sequencing process. To prevent beads from sticking in positions other than the intended bead locations, the flow can be increased to remove any weakly held beads, yet retain the correctly held beads. After a chemical process such as sequencing or amplification has been completed, the beads may be removed by either reducing the retention field flux, by adding a new field that pulls the beads away from the array, by increasing the flow rate of the fluid, by using the air water interface associated with an air bubble, which may include a surface tension force, or any combination of the above steps.

In an certain embodiments, as shown in Figure 3, an array of electrodes is used to retain charged beads, using either a DC field or a dielectrophoretic field, or both. As with the magnetic array, no well structure is needed to retain the bead, permitting free flow of components in solution. To insure that charged components in solution, such as the DNA sample, nucleotides, enzymes and other charged moieties may be readily able to flow through the volume above the array, a frequency of oscillation which is sufficient to retain the charged beads is used, but which is sufficiently slow as to permit

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 the moieties in solution to flow or diffuse away from a bead which is retained. The addition of depressions associated with the sensors in a one to one correspondence may result in better alignment between the beads and the sensors permitting better detection. In an alternative embodiment, pedestals or registration posts or other three dimensional structures are used, for example, for better fluid flow may be incorporated.

In an alternative embodiment, wherein the beads may be localized in a one to one correspondence to the array, the beads may be brought into position by a magnetic or electrical field, and may then be held in place by an alternative means, such as DNA hybridization, biotin streptavidin binding, thiol binding, photo-activated binding, covalent binding, or the like. The binding may be initiated by a change in temperature, application of light, or by washing in a binding reagent or catalyst whilst the beads are held in a one to one correspondence by said magnetic or electric field. After binding has occurred, the magnetic or electric field strength may be allowed to change in intensity or frequency, potentially being turned off. The binding may be reversible, permitting the beads to be washed out of the volume above the array of sensors.

Tn some embodiments, the magnetic or paramagnetic particle may have surface coating thereon that is of sufficient porosity to provide access for a polymerase or other enzyme, as well as for sample DNA, dNTPs ions and other moieties to pass through. The coating may be configured so that primers may be attached at an appropriate spacing, and may thereby provide a greater density of sample DNA and thus charge to interact with the sensors located thereby. Said coating may be of agarose, polyacrylamide or other cross linked polymer, or may be made of porous glass.

In other embodiments, the bead may have coatings configured so as to minimize or reduce nonspecific binding of DNA, proteins, or other charged moieties relative to the amount of nonspecific binding which may result when said DNA proteins, or other charged moieties interact with said beads without said surface coating. Said coatings may be similar to coatings described herein for use on the surface of a flow cell, sensor, enrichment module, or magnetic array, and any coating herein described for one surface may be utilized on other surfaces.

In some embodiments, the beads have a magnetic core, which may have an impermeable coating thereupon. Said coating may be bound, attached or associated with multiple strands of DNA. For example, the DNA strands may each be substantially identical rolling circle amplicons, providing multiple strands of DNA where each strand has multiple contiguous copies of one DNA target.

Figure 8 illustrates one embodiment for an alternative method and system for retaining beads in a one to one correspondence with the sensors in an array. Figure 8 illustrates a system 800 whereby individual control lines 810 are activated and thus one

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 layer of a structure 812 is expanded into the flow cell volume 806 between a substrate 804 and a fluidic structure 802 sequentially, forcing beads 808 out from a level exceeding a one to one correspondence with sensors, by displacing excess beads 808 as the control lines 810 are activated. The beads 808 may then be held in place during sequencing cycles. As can be seen in Figure 8, there is sufficient room for liquid flow below the control lines, but insufficient room for bead movement. When a set of sequencing cycles has been completed, the control lines may be deactivated, and the beads 808 may then be removed by fluid flow through the sensor array region.

In an alternative embodiment, the number of beads may be lower than the number of sensors. The number of beads may be close to the number of sensors, with control lines being activated such that beads are caused to be localized with sensors. Said localization may be assisted by alternating flow directions in the sensor array region, introducing vibrations or oscillations, or the like, such that the beads are undergoing frequent motion, until such time as motion is prevented by the control lines being sufficiently activated that said control lines arc too low to permit beads from being able to move from one sensor region to another. Further movement of the beads combined with further activation of the control lines will serve to more completely center the beads over the sensors.

In a further alternative embodiment, a structure having a shape similar to that shown in Figure 8 may be molded, machined, or otherwise formed such that the shape is similar to that which will occur wherein the control lines are fully activated. Said structure may be slowly lowered over the bead covered sensor array.

In yet a further embodiment, in order to cause a higher percentage of sensors to have associated beads, beads may be attached to sensors by biotin streptavidin binding thiol binding or the like after a set of beads has been localized by one of the aforementioned structures. Additional beads may then be introduced into the sensor array region, and the process repeated. If binding agents are localized to areas above sensors, as may be done as previously described, any excess beads which are caught or pinched by the structure will not bind, and may be washed away before beginning the sequencing cycles.

In yet a further embodiment, a significant excess of beads may be introduced into the sensor array region. A single control line at an exit from the sensor array region may be activated to trap the set of beads. The aqueous conditions and/or temperature may be changed to permit binding of beads to the sensors. The excess of beads may then be removed by releasing the control line allowing flow of the beads out of the sensor array region with an aqueous reagent.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Some embodiments combine pH sensing with electrochemistry detection as a result of the incorporation of a reversibly reducible layer which may be fabricated above the previous sensor design. Such sensors are available from Senova Systems. During a sequencing cycle, a reducing reaction will occur if a base has been incorporated in the bead associated with a sensor. The level of reduction can be measured, and after the completion of the sequencing cycle, a voltage can be impressed on the sensor, causing an oxidation of the surface, returning it to its original state, whereupon it can be used for the next sequencing cycle.

In some embodiments, magnetic beads are used without a magnetic array. The magnetic beads self assemble into a monolayer with uniform spacing, the spacing of which may be influenced by the use of an external magnet to change the local field strength. Said magnetic beads may be caused to be separated at a spacing which matches the spacing of a sensor array, and then may be caused to bind to the sensor array by changing aqueous conditions, temperature or the like as previously described. Slow translation or movement of the beads may be appropriate after binding in order to enable alignment of the beads with the sensors. Such translation or movement may need to occur in multiple dimensions, which may include x, y, theta, and spacing. The addition of depressions associated with the sensors in a one to one correspondence may result in better alignment between the beads and the sensors permitting better detection.

Figure 11 illustrates various embodiments, where the magnetic, paramagnetic, non magnetic particles may be of shapes other than spherical for use with either a sensor array with magnetic retention (1102), a sensor array with electrical confinement, or a sensor array with self assembled particles. The particles may be planar, round, rectangular (1104), star shaped, hexagonal (1106), or other shape. In other embodiments, the particle may dendritic, enlarging the surface area of said particle. Said dentritic particle may be generally spherical, planar, oval, or any other shape. In yet other embodiments, said particle may be porous; if said particle is porous, the pore size may be of sufficient size as to permit free movement of DNA, polymerase, dNTPs and other moieties necessary for primer extension sequencing or other applications as appropriate.

In another embodiment, not requiring a magnetic bar array, the beads may be replaced with DNA balls, created by using rolling circle amplification. The DNA balls may then be digested by a DNA nuclease. In another embodiment, the balls may fabricated from a monomer or polymer, such as polystyrene, which may be dissolved subsequent to sequencing, using an organic solvent such as acetone, thus freeing the attached DNA by the same process, both of which may then be removed from the flow cell by flowing a reagent through said flow cell.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In yet another embodiment, the bead may be held in place by a binding between moieties attached to the well and moieties attached to the bead. The well may be shorter than the radius of the bead, and may have a shape other than circular, so that reagents may flow around the bead, while the bead may be bound at several points at the entrance to said well. The binding may be a Streptavidin Biotin binding, a DNA DNA binding, a DNA PNA binding, a PNA PNA binding, Thiol Au binding, photoactivated binding, covalent binding, or the like. Said binding may be released by raising the temperature, or by introducing a reagent which reduces the affinity between the moiety on the well and the moiety on the bead, permitting the beads to be removed from the flow cell by flowing a reagent through said flow cell. The beads may be further induced to move from the well in which said bead has been bound by sonicating the bead and well structure. Sonication may be done at the same time that a reagent is flowed through the flow cell in order to remove said beads from said flow cell.

In the process of amplifying DNA in chamber free system as described in provisional application 61/491081, various factors may be potentially subject to optimization. Among these include the frequency, voltage and size and shape of the confinement “cell” used to confine the polymerase, target DNA and generated amplicons. If confinement were the only consideration, it would be possible to confine almost any size of amplicon, without regard to the small size of said amplicon. However, in order to be able to have a field strong enough to insure proper confinement, the field may prevent proper activity of the polymerase incorporation of bases during the PCR or isothermal amplification process, or may pull the polymerase and or extended primer from the complex of the target DNA extended primer and polymerase. In one embodiment, it is desirable to optimize a combination of frequency, voltage and size of the confinement cell, depending on the size of the amplicon.

In some embodiments, the amplification reaction is inverse PCR amplification, hot start PCR amplification, methylation specific PCR amplification (MSP), nested PCR amplification, reverse transcription reaction, reverse transcription PCR amplification (RT-PCR), Touchdown PCR amplification, intersequence specific PCR amplification (ISSR-PCR), co-amplification at lower denaturation temperatures (COLD-PCR), solid phase amplification, bridge PCR amplification, or single primer bridge amplification.

In some embodiments, the amplification reaction is helicase dependent amplification (HDA), a nicking enzyme amplification (NEAR), recombinase polymerase reaction (RPA), transcription mediated amplification (TMA), self-sustained sequence replication (3SR), nucleic acid based amplification(NASBA), signal mediated amplification of RNA technology(SMART), loop mediated isothermal amplification of DNA (LAMP), isothermal multiple displacement amplification(IMDA), solid phase

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 isothermal, bridge isothermal, single primer isothermal amplification (SPIA), circular helicase dependent amplification(cHDA), or rolling circle amplification.

In generating a DC field for electrophoretic concentration or confinement, electrolysis products can build up. These include hydronium and hydroxide ions. To minimize effects from these ions the DC field can be pulsed so the net DC is much lower. In some embodiments the pulse duty cycle can be reduced after the DNA has migrated closer to the center electrode. In other embodiments the process can use DC to concentrate the DNA and then use AC to maintain concentration or confinement of said electrolysis products. In other embodiments both DC and AC can be used for concentration and confinement.

In some embodiments, pulsed field gel electrophoresis is used. For example, a non-sinusoidal AC waveform may be used, where a higher positive voltage may be balanced by a longer but shorter negative voltage such that the average voltage is substantially zero. The higher positive voltage may be used with a polymer concentration, such that reptation of the DNA occurs in the polymer solution. The polymer solution may effectively cause a lower migration of DNA in the direction of the lower field compared to the migration in the higher strength electric field, which may thereby increase the mobility of the DNA. In this manner the DNA may migrate more in the desired direction while other molecules such as Mg move freely back and forth due to the balanced nature of the AC waveform. The migrational variance may also be frequency dependant, so that different sizes of DNA may be captured. The pulsed field gel electrophoresis may be ID or 2D, and may use contour clamped electric field, transverse alternating field electrophoresis, or rotating gel electrophoresis, any of which may be used with either a gel, an entangled polymer, or another sieving matrix.

In generating a dielectrophoresis field, typically a sinusoidal waveform is used. While this may be ideal for an application intended strictly for confinement or separation of different species, it may cause issues for a system where a biochemistry reaction may be performed within the confinement volume. For example, high fields are likely to result in localized heating. In one embodiment, a modified sinusoidal waveform may instead be used. For example, the modified sinusoidal waveform may have the voltage removed at the top of the sinusoid, or at any other point in the sinusoidal waveform, allowing localized diffusion, permitting hybridization of the amplicons to primers, binding of polymerase to the duplex DNA, and binding and incorporation of nucleotides or nucleotide analogs. The field may then be reinstated after an appropriate period of time. The same process may occur at the peak with the opposite sign in the modified sinusoidal waveform. In other embodiments, any other alternating current waveshape can be used for concentration or confinement. Alternatively, the interruption in the sinusoidal waveform may occur only once per

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 cycle, or may occur once in every several cycles, or once in many cycles, so that any “stray” amplicons may be captured in the regions with lowest field strength and returned to the main volume of the confinement volume. Alternatively other wave forms such as square, trapezoidal, non symmetrical wave forms, etc may be used.

In some embodiments monoclonal beads may be generated in a small microfluidics device. In one embodiment the electrodes and magnets may be fabricated on a thin sheet wherein a top surface may be bonded or fused in place creating an integrated microfluidics device. The microfluidics device may have electrical and fluidic connections made to it. The microfluidics device may then be placed in good thermal contact against a first heated plate, for example by vacuum or air pressure. A second plate may be situated above at a different temperature. The two temperatures may be chosen to facilitate PCR amplification. After one temperature point is complete the card may be transferred or caused to come into contact with the upper heated plate, for example by vacuum or air pressure. Because only the thin card and the reagents in the card need to change temperatures the system can have fast temperature transitions, and consume minimal power.

In other embodiments that provide a minimal thermal mass, the electrodes built into the amplification microfluidics device may be used as resistive heaters to locally heat the liquid. In some embodiments the change in resistance of the electrodes are used to measure the temperature for better thermal control. In other embodiments the sensors, such as the NanoBridge or NanoNeedle (described herein) are used as a temperature sensor for better control of the area of interest.

In some embodiments, it may be desirable to perform DNA amplification from a single copy of DNA. If a polymerase error is made in amplification at an early point in the amplification process, such as in the first cycle in a PCR amplification, the error will proliferate such that it may not be possible to differentiate between the correct sequence and the error sequence. Thermostable polymerases typically have much higher error rates than mesophilic polymerases or thermophilic polymerases which may be not suitable for PCR as they inactivate during the denaturation step of PCR. Thus in some embodiments, it is desirable to use a highly accurate polymerase for the initial portions of a PCR reaction, where the highly accurate polymerase is not sufficiently thermostable to prevent inactivation during PCR, but may provide better accuracy than a more thermostable polymerase. The highly accurate polymerase may have a low Kotr, such that the highly accurate polymerase is substantially bound to the active extension site, in the presence of other polymerases which may be sufficiently stable to prevent significant inactivation during the denaturation steps of a PCR amplification.

In some embodiments, the highly accurate polymerase is introduced to a volume with primers and template prior to the introduction of the more thermostable

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 polymerase. In other embodiments, the thermostable polymerase is heat activated, such that any heat activated polymerase will be inactive for the first cycle(s) of the PCR.

In other embodiments a combination of isothermal and PCR amplification reactions is used. Initial amplification may be performed by a highly accurate non thermal stable polymerase, and subsequent amplification may be performed by a less accurate thermal stable polymerase that is not substantially inactivated by the denaturation step of PCR.

In an alternative embodiment, a clonal population is generated in the area of individual sensors in a sensor array. The sensors may be NanoNeedles or Nanobridges or other sensors to detect the event of polymerization. In one embodiment, primers are preferentially attached to the surface of the sensors. The primers may be preferentially attached as a result of a difference in materials, where the material of the sensor is more advantageous for attachment then the areas between the sensors of the sensor array. In an alternative embodiment, a mask may be applied to areas between the sensors of the sensor array, and a surface modification may then be performed. Subsequently, the mask may be removed; leaving an area between the sensors of the sensor array wherein the surface modification has not been performed. The surface modification may include attachment of biotin, applying a layer of gold and various other methods as are known in the art.

Primers may then be preferentially applied to the areas on the surfaces of the sensors in the sensor array. In one embodiment, the primers are attached as a result of a biotin streptavidin binding, where the streptavidin is attached to the 5’ end of the primers. In another embodiment, a thiol group may be attached to the 5’ end of the primers, which can then bind to the gold layer previously applied above the sensor, forming an Au-S bond. If a PCR reaction is desired, the primers may be modified with DTPA such that two thiol gold bonds are formed, preventing the dissolution which may otherwise occur from the 60 to 95C temperatures routinely used in PCR.

After a set of sequencing cycles is completed, the primers are removed and replaced. Buffer conditions can be changed to weaken the biotin streptavidin bond, such as high concentrations of GuHCl at low pH; alternatively the temperature can be raised to over 70 C to break the biotin streptavidin bond. Thiol bonds can likewise be broken at elevated temperatures. Aggressive means may be used, as damage to the polymerase and DNA is no longer consequential. In one embodiment, organic reagents arc used to break the binding between the extended primer and the surface, such as a covalent binding. After the extended primers are removed, new primers may be flowed into the volume above the sensors, enabling the device to be used again for another set of sequencing cycles on a another set of DNA samples.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments as schematically depicted in Figure 12, the sensor array may have elements 1200 that are provided with an additional array of electrodes 1206, which may be used to perform dielectrophoretic concentration. Dielectrophoretic concentration may be initially performed to attract sample DNA dNTPs, and primers to each sensor region. Amplification can then commence in the region of each sensor where a sample DNA molecule is located. During the amplification process, the dielectrophoresis forces may also aid in preventing cross contamination between different sensor regions undergoing amplification by retaining amplicons. In order to insure that polyclonal regions are not generated, the concentration of input DNA needs to be low enough that most sensor regions have one or zero sample DNA molecules. DNA samples could be single stranded or double stranded depending on the amplification methodology. The amplification reaction can be either a PCR reaction, or an isothermal reaction. In some embodiments the additional electrodes 1206 arc shown as having the same voltage relative to voltage level of the sensors. In an alternative embodiment electrodes on either side of a sensor may have voltages of opposite sign relative to each other.

Said sensor array elements 1200 may be fabricated on a substrate 1212, and may have magnets 1216 utilized for retention of magnetic or paramagnetic particles or beads 1202, wherein said magnetic or paramagnetic particles or beads may be held down against an electrode 1204, an against a dielectric 1210 and or upper electrode 1208. Detection may utilize said electrode 1204 and said upper electrode 1208, while dielectrophoretic concentration/confinement may utilize electrode 1204 and outer electrode 1206, wherein said outer electrode may comprise a single electrode or may comprise multiple electrodes.

Amplification may be a solid phase amplification, where one primer is on the surface of the bead, and a second primer is in solution, or the amplification may be solid phase where all primers are on the bead. Alternatively, the amplification may be performed where both primers are present in solution, and one primer, or both primers are also present on the bead.

The electrode configuration may take various different forms, including a planar electrode on both major planes of the flow cell, or there may be one electrode on the surface opposite the beads, and a set of smaller electrodes associated with each detector location.

Figure 12 illustrates the use of the amplified regions above the sensors in the array of sensors in a sequencing reaction. After the amplification reaction has been completed, the volume above the sensor array may be washed, removing amplicons, polymerases, and dNTPs. Polymerase and individual dNTPs may then be flowed into the volume above the sensor array, permitting binding, incorporation, and detection of 56

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 the incorporation events, resulting in the determination of the sequence of the different amplified sample DNA molecules.

In some embodiments, the sensor is used for several purposes, such as, for example, detecting the presence of beads when introducing beads, detecting amplification associated with the bead (e.g., real time PCR amplification or end point PCR amplification), and detecting sequencing reactions.

When generating clonal beads a large percentage of the beads will have no DNA template. In addition others may have poor amplification. These beads do not provide useful sequencing so it is desirable to remove these beads to improve instrument throughput and reagent utilization. In some embodiments, beads with no or minimal amounts of template are separated using an electric field. The beads on which amplification has occurred have more fixed negative charge from the amplified DNA and can be separated from those, on which amplification has not occurred, by the use of an electrophoretic separation. This permits the situation as shown in Figure 3 where most positions in the magnetic array are depicted as being occupied by beads which have had an amplification reaction, and are therefore suitable for use in a sequencing reaction.

Beads fully loaded with templates have a higher charge so will move farther in an electric field than beads with only primers or few templates. In one embodiment as shown in Figure 13A, Figure 13B and Figure 13C, this separation is done in a flow through module. A first fluidic input 1311A allows the injection of mixed beads. A second inlet 1312A allows the injection of a buffer solution without beads. A first outlet 131 IB is downstream from the first inlet 1311 A. A second outlet 1312B is downstream from the second inlet 1312A.

The fluidic flow rates can be set by fluidic resistance or pumping speed such that more liquid flows in the second inlet. In the embodiment shown in Figure 13B, the inlet and outlet widths may be varied to create different fluidic resistances, but other methods of modifying the fluidic resistance such as different length or height may be anticipated, or the use of flow restrictors in parts of the system external to the enrichment module 300. Similarly the fluidic resistance of the first outlet 131 IB and second outlet can be modified so more liquid flows out the first outlet 131 IB. In such a setup beads without a small velocity perpendicular to the flow will exit the first outlet port 131 IB. Additional output channels can be added to facilitate separation of beads with medium levels of template. In some embodiments, the flow rate in each output channel may be controlled directly by providing individual pumps for each outlet channel.

A pair of electrodes 1313, may be provided which enable generation of an electric field perpendicular to the fluid flow in the separation section 1310 such that the

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 template loaded beads, which may be brought into the enrichment module 1300 through inlet 1311A while an additional reagent may be introduced to the module through a second input 1312 A, may migrate out of the flow path towards second outlet 1312B. Fluidic ports 1309 allow connection to the system plumbing.

The electrodes 1313 could be made of any electrode material compatible with electrophoresis. In some embodiments discrete metal wires may be used but metal traces may be also anticipated. Metals such as platinum, platinum/iridium, gold and other noble metals or alloys may be anticipated as well as corrosion resistant materials such as stainless steel. Non metal electrodes may be also anticipated.

In another embodiment the fluidic flow rates can be set by fluidic resistance or pumping speed such more or less fluid flows out the first outlet than the second outlet allowing steering of the bead flow stream. In yet another embodiment the bead stream could be adjusted by both the inlet and outlet fluid flows.

The flow through enrichment module 1300 can be constructed of non conducting materials such as molded plastic, glass, ceramic or moldable polymers such as PDMS, or of conductive materials which may be coated with a nonconductive coating, or combinations or these materials, or in combinations with other materials. Fluidic components can be fused, bonded, or held together with a clamp mechanism to create an enrichment module including a separation section 1308. The enrichment module may in one embodiment include a molded upper piece 1308, and a flat substrate 1302. In other embodiments, the enrichment module may be made of more than two pieces, such as for example, three, four, five or more components. If two components may be used, both sides may have non-planar surfaces, such that fluidic or control channels may be formed in either component. If more components are used, any one of them can be planar or shaped such that they include channels, depressions or protrusions, or may be a combination of planar and shaped such that they include channels, depressions or protrusions.

In some embodiments, the surface or a portion of the surface of one or more components of the enrichment module has a zeta potential sufficient to cause significant electroosmotic flow. It may be desirable to minimize any mixing or turbulence which might otherwise result from said electroosmotic flow. In some embodiments, materials such as TiO², ZrO², or BaTiO³ are selected such that the zeta potential and the resulting electroosmotic flow are significantly reduced. In some embodiments, the zeta potential and the relationship between the zeta potential and a change in pH may change depending on a surface coating. In some embodiments the zeta potential may change significantly with changes of pH, as is the case with the pH dependency of Silica. In other cases, the zeta potential changes very little with respect to pH, particularly in the pH range from pH 7.5 to pH 9, as is the case with the pH dependency of BaTiO³.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In other embodiments, surface coating(s) such as PEG (Poly Ethylene Glycol), methyl cellulose, n-dodecyl-B-D-maltoside, acrylamide, fluorinated alkane chains, PTFE, acrylate, or other cross-linked or partially cross-linked polymers are used to modify the zeta potential, or combinations of surface coatings are used to similarly minimize the electroosmotic flow. In some embodiments, polymers are used to fill the aqueous volume of the electroosmotic flow restriction section.

In other embodiments, a physical structure as shown in Figure 13C is used to reduce or eliminate unwanted mixing and turbulence due to electroendosmotic flow. Such a structure may have a flow constraining section 1320, and said flow constraining scction(s) 1320 may be used on both sides of separation section 1308. Electric field may be distributed from electrodes (not shown) through buffer reservoir section(s) 1322. Said electrodes may be positioned in auxiliary input port(s) 1324, which may be used to bring buffer to and through the buffer reservoir section(s) 1322 and flow constraining section(s) 1320. In an alternative embodiment the electrodes are positioned in the buffer reservoir scction(s) 1322, with electrical connection to a voltage source which may be external to enrichment module 1300. Input beads and reagents may be brought into said separation section 1308 through input ports 1311A and 1312A, or reagents may be brought in through input ports 1311A and 1312A, while beads are brought in through center input port 1326. Beads may be separated between output ports 131 IB and 1312B.

In an alternate embodiment the electrodes are positioned in a buffer reservoir(s) separate from the structure shown in Figure 13C with a fluidic connection to allow current flow into the enrichment module. In some embodiments, the buffer reservoir(s) allow electroendosmotic flow to more in one direction across the enrichment flow cell, rather than having a circular flow with concomitant turbulence that would occur with a sealed flow cell, where any flow in one direction must be matched by flow in the opposite direction. In some embodiments the voltage can be stopped periodically to allow the fluid reservoirs to return to their equilibrium state, where the liquid level in each reservoir is at the same level, after electroosmotic pumping. In other embodiments, the volume or cross section of said reservoirs is significant relative to said electroosmotic pumping, such that bead sets or multiple bead sets may be separated or enriched prior to allowing said fluid reservoirs to return to their equilibrium state.

In other embodiments the zeta potential magnitude is reduced by protecting the silanol groups with a compound such as trimethylchlorosilane which decreases the number of ionizable silanol groups. Figures 13D and 13E microphotographically illustrate the use of a structure as shown in Figure 13C, where a stream of input beads 1314 is illustrated without a field applied in Figure 13D, and thus all input beads 1314 are carried to output 131 IB as if they were lower charged beads 1314A, and no beads

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 appear to be higher charged beads 1314B pulled to outlet port 1312B. The flow constraining sectionl320 shown in Figure 13C is shown in more detail, showing the fluidic passages 1316, and support columns 1318. In Figure 13E, where a field is applied to said separation section 1308, and said stream of input beads 1314 is separated into low charged beads 1314A which is carried to output port 1311 A, and higher charged beads 1314B are pulled and carried to output port 1312B.

In some embodiments, the thickness or depth of the separation section 1308, fluidic inputs 1311 A, 1312A, fluidic outputs 131 IB, 1312B, separation section 1308, electroendosmotic flow restriction section(s) 1320, and buffer reservoir(s) 1322 may be the same. In other embodiments, the thickness or depth may be different for different sections, for example, the thickness of the electroendosmotic flow restriction section(s) 1320, may be less than that of the separation section 1308 or the buffer reservoir(s) 1322.

In some embodiments, the thickness or depth of the separation section 308, and other fluidic sections of the enrichment module is from 10 to 1000 gm; in other embodiments the thickness or depth of the enrichment module is from 20 to 200 gm, 50 to 150 gm, 200 to 500 gm, or from 70 to 130 gm.

In some embodiments the width of the flow restrictors in the electroendosmotic flow restriction scction(s) 313 is from 10 to 1000 gm, in other embodiments, the width of the flow restriction is from to 20 to 200 gm, 50 to 150 gm, 200 to 500 gm, or from 70 to 130 gm.

In some embodiments the length of the enrichment zone can be up to 2mm, 2mm to 10mm, 10mm to 100mm. In some embodiments the width of the enrichment zone can be up to 1 mm, 1 mm to 4mm, 4mm to 10 mm and 10mm to 100mm.

In some embodiments, the enrichment module 1300 has a feedback system (not shown) to compensate for varying charge levels that may exist between different batches of beads. Such a feedback system may then permit the electrophoretic voltage to be automatically tuned for a particular batch of beads, and to be automatically retuned for a subsequent batch of beads. Said feedback system may use reflected light, absorbed light, deflected light, fluorescent light, capacitive coupling to the beads, direct conductivity detection of the Debye layer associated with the beads, ISFET/ChemFET detection of the beads, or may use any other appropriate detection means. Said detection means may be configured such that detection of beads is effectuated in or associated with one or more fluidic outputs 131 IB and 1312B, or may be configured such that detection of beads is effected in or associated with the separation section 1308.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments the feedback system is used during a separation of a batch of beads, tuning the flow rate and electoosmotic voltage so that the beads are optimally separated and flow into the nominal desired position in each output 131 IB and 1312B. Said flow rate control may be effectuated by the use of a variable pressure applied to said flow, a variable vacuum applied to said flow, a variable restriction in the said flow, or any combination thereof.

In some embodiments, the bead slurry may need to be concentrated. In one embodiment the bead solution is passed by a magnet to hold the beads. By removing the magnetic field or using a higher flow rate the beads can be released in a more concentrated form.

In some embodiments the enrichment module 300 is used to separate negatively charged DNA from proteins, including cellular membrane fragments which may be comingled with said DNA after lysing of the cell(s) from which said DNA and said proteins and said cellular membrane fragments may have originated. In some embodiments it may be desirable to separate the DNA from proteins, most of which may be positively charged, and to also separate the DNA from proteins which may be negatively charged at neutral pHs, such as human scrum apotransferrin, thyroglobulin, or BSA. Such proteins which may be negatively charged at neutral pHs typically have pKa values above 4.0, whereas the pKa for DNA is 1.0. The electrophoretic mobility of proteins is typically much lower than that of the highly negatively charged DNA, permitting easy separation of DNA in an enrichment module 300. Such separation may be performed at a low pH such as a pH below 7, a pH between 6 and 7, a pH between 5 and 6, a pH between 4 and 5, or a pH between 3 and 4, permitting the enrichment module to run below the pKa of said proteins, and above the pKa of the DNA.

In some embodiments, the DNA may be dielectrophoretically captured after being substantially separated from any proteins and cellular membranes. Said dielectrophoretic capture may be effected in a fluidic outlet 31 IB, 312B, or may be effected in a separate module. After said dielectrophoretic capture, the buffer may be changed, for example from a low pH as previously described to effect a separation from proteins or cellular membranes, to for example, a buffer suitable for PCR, isothermal amplification, or sequencing, where the pH is approximately optimal or otherwise suitable, for example, for polymerase activity.

The voltage applied to the electrodes 1313 can be reduced or even reversed periodically if necessary should beads stick to the electrodes. The voltages used may be greater than that required for electrolysis (1.23V at 25C at pH 7), or may be less than the voltage needed for electrolysis. Higher voltages and narrower gaps provide a higher field strength and more force on the beads. The voltage on the system can be calibrated by flowing beads without or with limited template and setting the voltage and or flow

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 rate such that these beads may be not moved far enough to enter the second outlet while beads with template may be directed into the second outlet.

Non-flow-through enrichment modules may be also anticipated. In one embodiment beads are introduced to a chamber and a magnetic field or gravity pulls the beads down. An electric field is established pulling the beads with template up. In some embodiments a capture membrane or filter can be added in front of the positive electrode to facilitate concentration of the beads.

In one embodiment, beads are removed from the flow cell as a result of actions and methods which are performed within the same instrument where the flow cell is used to sense a reaction, such as a sequencing reaction.

In another embodiment, the flow cell assembly is removed from the instrument where the flow cell is used to sense a reaction, such as a sequencing reaction, and moved to another instrument or device, where the beads are removed from the flow cell.

In yet another embodiment, the flow cell assembly is removed from the instrument where the flow cell is used to sense a reaction, such as a sequencing reaction, and shipped to a central refurbishment site, where the beads arc removed from the flow cell.

In another embodiment, the flow cell assembly is removed from the instrument where the flow cell is used to sense a reaction, such as a sequencing reaction, and moved to another instrument or device, where coatings may be applied, removed and/or reapplied to the flow cell and/or fluidics manifold.

In yet another embodiment, the flow cell assembly is removed from the instrument where the flow cell is used to sense a reaction, such as a sequencing reaction, and shipped to a central refurbishment site, where coatings may be applied, removed and/or reapplied to the flow cell and or fluidics manifold.

In some embodiments, the flow cell and/or the fluidics manifold have various surface coatings. Such surface coatings are used to reduce nonspecific binding of moieties in the various reagents, to the surfaces in said flow cell or fluidics manifold. In some embodiments, the coatings intended to reduce nonspecific binding may include PEG (Polyethylene Glycol), BSA (Bovine Serum Albumin), PEI (Polyethylenimine), PSI (Polysuccinimide), DDM (n-dodecyl-b-D-maltocide), fluorinated coatings, Teflon coatings, silanization coatings, or other appropriate coatings.

In one embodiment, the coatings arc applied, removed and/or reapplied to the flow cell and or fluidics manifold as a result of actions and methods that are performed within the same instrument where the flow cell is used to sense a reaction, such as a sequencing reaction.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments the sensor combines pH sensing with electrochemistry detection as a result of the incorporation of a reversibly reducible layer which may be fabricated above the previous sensor design. Such sensors may be available from Scnova Systems. During a sequencing cycle, a reducing reaction will occur if a base has been incorporated in the bead associated with a sensor. The level of reduction can be measured, and after the completion of the sequencing cycle, a voltage can be impressed on the sensor, causing an oxidation of the surface, returning it to its original state, whereupon it can be used for the next sequencing cycle.

As shown in Figure 9, in some embodiments, a redox reaction may be performed where the redox potential may comprise a combination of an AC potential 904 combined with a nominal DC potential 902, where the nominal DC potential 902 is one half of a sine wave, starting at zero volts, rising to a maximum, and returning to zero volts, where said AC potential 904 may be superimposed on said nominal DC. The AC potential 904 waveform may be of a frequency which is 10 times as high as the nominal DC potential 902, or AC potential 904 waveform may be of a frequency which is 10 to 100 times as high as the nominal DC potential 902, or AC potential 904 waveform may be of a frequency which is 100 to 1000, 1000 to 10,000, 10,000 to 100,000, 100,000 to 1,000,000, 1,000,000 to 10,000,000 times as high as the nominal DC potential 902. The AC potential 904 waveform may be a sinusoidal waveform, a triangular waveform, a square waveform, or any other sort of symmetrical or asymmetrical waveform. The DC potential waveform may be half of a sine wave, an isosceles triangular waveform, a saw tooth waveform, or any other waveform starting at zero volts and increasing to maximum and thence returning to zero volts. The magnitude of the superimposed AC potential 904 waveform may be constant, or may change during the DC potential 902 waveform, for instance, the AC waveform may be smaller when the DC waveform is close to zero, and the AC waveform may increase as the DC waveform reaches its maximum potential. The current 906 resulting from the combination of the AC potential 904 and nominal DC potential 902 may be a non linear function of applied potential.

In one embodiment where the quantity of data which is collected is minimized, it may be desirable, in addition to increasing the speed of the reaction, to align the active regions with the normally occurring rectilinear nature of semiconductor electronics. In previous systems, the locations of the active reactions may not align well with an array of detectors. It is preferable to arrange the detector electronics in a strict rectilinear fashion, as opposed to convention which brings reagents in and out from the comers 1404A and 1404B of a chip 1400 as shown in Figure 14. This approach both prevents alignment with the reagent slug, and wastes a significant area of the chip, as the reagent slug cannot flow well (or at all) to the other corners of the chip 1402, as well as causing nonuniformity of flow (and loading efficiency 1406) due to the large 63

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 difference in cross sectional area from the center of the chip to the inlet and outlet ports in the comers of the chip 1404A and 1404B.

In one embodiment, multiple fluidic inlets 1512A as shown in Figure 15 may be used in a multiple flow cell sensor device 1500, permitting greater utilization of the chip area, and more uniform and aligned reagent flow relative to the chip and said chips readout structure. In some embodiments, samples may be introduced at different times to the different channels of the flow cell, permitting different samples to be used without the need for bar coding or other means for sample identification. The multi flow' cell sensor device may have multiple flow' cells 1510, with an input port 1512A an output port 1512B, and a line to waste 1516, for each flow cell, and wherein each input and output port may be configured so as to optimize uniformity of flow across the sensor array within said flow cell 1510, potentially having angled side walls and or surface roughness so as to optimize said flow uniformity. Said flow cells may further be configured with valves 1504 and controls for said valves 1506 for controlling the flow of samples and or other reagents in immediate proximity to the flow cells 1510. The samples and other reagents may be brought in using input ports 1502.

in some embodiments, the samples arc introduced different times in a process or set of processes. For example, a second set of beads that have undergone an amplification reaction and consequently have extended primers are introduced to a channel of a flow cell associated with a chip, where another channel of the chip may already have a first set of beads. The first set of beads may have already undergone an amplification reaction and consequently have extended primers contained therein, and the channel with the first set of beads may have been exposed to a sequencing cycle, or may have been exposed to a number of sequencing cycles. In another embodiment, the second set of beads is introduced to the same channel where the first set of beads is contained. In another embodiment, wherein amplification and sequencing may be performed in a single area of an array as described elsewhere herein, a set of beads in one channel may be undergoing an isothermal amplification, while a second set of beads in another channel may be undergoing a sequencing reaction.

In some embodiments, wherein a temperature change may be needed for a first process but not for a second process, the second process may be temporarily halted or paused while the temperature change occurs, and may then commence after the temperature is returned to the previous temperature. For example, an amplification reaction process may need to melt off and remove a second strand after primers associated with a bead have been extended, so that complementary unextended primers can be hybridized to the primers associated with the bead, so that a sequencing by synthesis reaction may commence. Prior to raising the temperature of the chip with multiple channels, the channel having a set of beads undergoing a sequencing reaction

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 may pause the sequencing reaction, and may actuate the co-localized virtual nanoreaction well electrodes. In this manner, hybridized partially extended primers are kept localized with each virtual nanoreaction well, while the extended primers not associated with beads in the channel, may be dissociated from the primers associated with the beads, and then removed from the channel.

The system could divide the samples over multiple fluidic channels or chips if they are too large, or combine the samples if they are combinable (for example barcoded samples). In some embodiments samples provided to the instrument would be ready for sequencing. In other embodiments samples could be processed by the instrument to generate sequencing ready samples.

In another embodiment, multiple input ports to a single flow cell or electrowetting arc used to introduce samples to portions of a flow cell, permitting more samples to be used at a time, without risk of cross contamination.

in another embodiment, an clcctro-wctting system may be provided to move reagents within a part of the flow cell without generating a reagent interface, thus completely preventing any mixing of reagents prior to the reagent reaching a desired portion of the flow cell, and may thus provide an extremely sharp transition in reagents.

In some embodiments, it is desirable to have a sharper transition between one reagent and another, for example, when flowing a dNTP containing reagent through the flow cell where the sensors are positioned. This may help to provide a quicker transition between a concentration where essentially zero dNTPs are present, to a reagent a concentration at a desired dNTP for incorporation by polymerase to extend, or further extend the primer. This may enable a shorter time between a time of initiation of base incorporation and a time where a significant percentage of the primers or extended primers have had the newly introduced nucleotides incorporated into appropriate locations in different colonies in the flow cell. The shorter time may allow a greater change in the signal level of a sensor or sensors per unit time, which may improve the signal to noise. Electronic sensors are typically subject to having noise, which may include thermal noise and flicker noise, both of which can be minimized by reducing the time interval during which an integrated signal is sensed by a sensor.

This quick transition is drawn in contrast with a system where a slower transition occurs between a concentration where essentially zero dNTPs are present, to a reagent with a concentration at a desired dNTP for incorporation by polymerase to extend, or further extend the primer. This slower transition may occur as a result of diffusion of dNTPs from a reagent solution containing dNTPs into a reagent solution which initially has essentially zero dNTPs. This may occur while transitioning through a long channel through which reagents are flowed prior to introduction into a flow cell

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 where sensors are positioned for the detection of a sequencing reaction or another reaction. Further mixing may result from changes in channel width, comers through which the reagents are flowed, irregularities in the surface of a channel through which reagents arc flowed, slow flow through channels through which reagents arc brought to a flow cell, or slow flow through a flow cell.

In one embodiment, the channel length may be substantially reduced between a point in the fluidics system where a transition between a concentration where essentially zero dNTPs are present, to a reagent with a concentration at a desired dNTP for incorporation by polymerase to extend the primer is generated, and the point in a fluidics system where the reagents arc introduced into a flow cell. Said channel length between the generation of a transition between said reagents and said flow cell may be a micron or less, one to five microns, five to 20 microns, 20 to 100 microns, 100 microns to 300 microns, 300 microns to one millimeter, one millimeter to three millimeters, three millimeters to ten millimeters, or ten millimeters to 30 millimeters. Tn general the shorter the distance between a position in the fluidics system wherein a transition between said reagents is generated and a position wherein said reagents are introduced into said flow cell, and the fewer the number of transitions in the size of flow cross section of the fluidics system, or comers in said distance, the lower the diffusion of dNTPs or other moieties will be, and thus the shorter the time frame will be between a reagent with a concentration of essentially zero dNTPs and a reagent with a reagent with a desired concentration of dNTPs at each sensor in a flow cell.

Minimizing the time for the polymerization reaction may also improve the signal to noise associated with the detection of the polymerization reaction. The noise associated with the detector may be fairly consistent with time, and the total integrated amount of signal generated may be the same regardless of the period of time over which it is generated. Thus reducing the time over which the data may be taken by speeding the time associated with binding of the polymerase and/or by providing a quick transition in the concentration of the dNTPs as they are introduced in a reagent slug can minimize the amount of noise which must be dealt with in analysis. As a result, the noise bandwidth which must be dealt with by the analysis software may be reduced.

In some cases, where the sensor is near an exit port, the number of dNTPs needed for incorporation may be sufficiently large that depletion of dNTPs may result in an increased time to generate the desired dNTP concentration for incorporation by polymerase to extend the primer. Lower dNTP concentrations, longer distances with more colonies in a flow cell, larger colonies with more primers for extension may all result in an increased time to generate sufficient dNTP concentrations. In some embodiments, multiple input ports are used to provide inputs to a single flow cell, to ensure the availability of dNTPs at each sensor and/or colony. In some embodiments,

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 reagent channels are provided above the flow cell, in additional layers of a PDMS liquid manifold. Duplicate sets of control lines, similar to those shown in Figure 42 may be provided, which may control duplicate sets of valves, similar to those shown in Figure 42 and flow through duplicate sets of manifolds, similar to those shown in Figure 42, which may then provide an alternative input to the flow cell with a short distance between the point in the fluids where an interface between two reagents is created, and second or subsequent input port to the flow cell.

A significant issue associated with next generation sequencing is the enormous quantity of data generated. Some systems can generate an average of3000 or more data points for each useful base of sequencing data. Storing and analyzing data adds significantly to the overall cost of next generation sequencing. In some embodiments, data reduction is performed in the simplest way, by acquiring less data. Polymerase activity can be significantly more rapid than the time needed to bring reagents with dNTPs completely through a flow cell; thus DNA colonies close to the inlet of a flow cell may have completely finished the next synthesis before the dNTPs have even reached the colonies near the outlet of said flow cell. If data is acquired for the entire flow cell during the time needed to detect reactions occurring anywhere in the flow cell, much of the data will be from regions of the flow cell where no reaction is occurring. Depending on the time needed for the dNTP reagent slug to traverse the flow cell, and the speed of polymerization, most of the colonies in the flow cell will be either waiting for dNTPs, or will have completed their synthesis reaction, rather than incorporating dNTPs and thus producing useful data.

In some embodiments, the readout of the detector electronics is synchronized with the movement of the reagent slug through the flow cell. A reagent slug containing dNTPs may initially enter the flow cell, but not yet move far enough into the flow cell for the dNTPs to bind and incorporate with any of the colonies. At this point in time it may be possible to not take data at all. At a point slightly later in time the reagent slug will have entered the flow cell sufficiently to interact with the set of colonies in the first region. At this point in time data may be taken from the detectors associated with the colonies in a first region, but may be not taken for other regions of the flow cell. At a second point later in time the reagent slug may have entered the flow cell sufficiently to begin to interact with the set of colonies in a second region. At this point in time data may be taken from the detectors associated with the colonies in the second region, and may likely need to still be taken from the first region, depending on the speed of the reagent slug and the speed of the polymerase, but may not need to be taken for other regions of the flow cell. At a third point later in time the reagent slug has traversed through the flow cell sufficiently to begin to interact with the last set of colonies in the flow' cell. At this point in time data may be taken from the detectors associated with the colonics in the last region. Some data may still need to be taken from previous regions, 67

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 such as region immediately preceding said lat set of colonies depending on the speed of the reagent slug, the speed of the polymerase, the length of the flow cells, and the size of the colonies, but may not need to be taken for other regions of the flow cell.

In other embodiments, time multiplexing with phase delay may be performed to distinguish different samples from each other.

As the speed of valves used in a system may vary, and the size of tubes, channels and ports may vary from system to system and consumable to consumable, the flow rate may also vary. In order to accommodate said variation, the first set of data taken from a system or consumable for which prior data may provide appropriate guidance as to the rate of flow in the system, more data may need to be taken, so that it may be assured that the data associated with a sequencing reaction is captured.

In one embodiment the timing of collection can be adaptively determined from an earlier column, or from data from a previous cycle of the same column. For example, if the typical detection event is occurring near the end of the collection time an additional delay may be added before the start of the next collection period of downstream samples. Similarly if the typical detection event is occurring near the start of the collection time the next collection period for downstream samples may be started earlier.

The size of the regions is shown in the illustrations as being one colony wide, but the width of a region may be more than one colony wide. For example, if a flow cell is 1000 colonies wide from the inlet to the outlet, a region may be a one colony wide, 10 colonies wide, 100 colonies wide, 500 colonies wide, or any number in between. Thus the region from which data is acquired at a time as the reagent slug moves through the flow cell may vary from one sensor, to tens of sensors, to hundreds of sensors wide, moving with the reagent slug as it traverses the flow cell, and as the polymerization reaction completes.

The width of the region being read can be done at the time of reading the chip, preventing the need to generate and discard data. This can be done in a manner similar to that which may be used for reading a subarea of a CMOS image sensor, whereby a subset of the total rows or columns can be read out at a time. Depending on the device structure, it may be possible to select individual sensors, as it is possible to select individual pixels in some CMOS sensors. Alternatively, if the chip structure is designed to read out a complete row at a time, using separate analog to digital converters for each column of the array of sensors; the chip may be read out selecting which subset of rows are desired. The subset of rows will change as the reagent slug progresses through the flow cell, and as areas of the flow cell complete the polymerization reaction at the colonies in said area. In some embodiments, said

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 separate analog to digital converters of comparators associated with each column, wherein a counter may also be associated with each column, thereby allowing simultaneous conversion of the analog signal into digital signals, while allowing more time, potentially multiple orders of magnitude more time for the analog to digital conversion, with attendant improvements in signal to noise. In other embodiments, the analog to digital converters may be successive approximation devices. In other embodiments, a pixel parallel readout approach may be utilized.

The width of the data region being collected needs to be large enough to account for a variety of factors to insure that all valid data is taken. These factors include variations in flow rate of the reagent slug, which may have slower flow near the edges of the flow cell due to interactions with the surface. Other factors can include variations in the polymerization speed due to concentration of the polymerase, variations in concentration of the dNTPs, temperature variations, colony density variations, the number of repeats of the base being incorporated for a colony or colonies, amongst others. Any of these may require the width of data which is being taken to be longer.

As described previously, dNTPs for extending the primer may be native dNTPs, modified dNTPs which arc incorporable by a native or modified polymerase, or both native dNTPs and modified dNTPs. If a modified dNTP is used, the modification may act as a reversible terminator, a virtual terminator, or may change the charge of an incorporated nucleotide for easier detection. Thus in some embodiments, the sequencing reaction incorporates all of the bases in a homopolymer run, or may incorporate one base at a time in a homopolymer run, reducing difficulties in determining the number of bases in a homopolymer run when the number of bases in a homopolymer run is large.

The kinetics associated with diffusion and binding of the polymerase to the colony DNA may be noticeably longer than the kinetics associated with the diffusion, binding, and incoiporation of the dNTPs. As a result, the time period over which polymerization occurs may be longer if the polymerase is brought in the same reagent slug with the dNTPs, as compared with bringing in a reagent slug with polymerase, followed by bringing in a reagent slug with dNTPs. Thus it may be advantageous to the effort of minimizing the amount of data, to bring the polymerase into the flow cell, permitting the polymerase to bind to the colony DNA, prior to bringing in the dNTPs. If the polymerase is a processive polymerase such that said polymerase is well retained between cycles, polymerase may be combined with the dNTPs to eliminate the need for separate deliveries. In another embodiment, the delivery of a wash which includes phosphatase and polymerase may allow' effective elimination of residual nucleotides and replenishment of the polymerase in a single fluidic delivery. In other embodiments a reagent includes phosphatase without polymerase to effectively eliminate dNTPs by

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 removal of the phosphate group via hydrolysis. The phosphatase may be shrimp alkaline phosphatase, calf intestinal phosphatase, or another phosphatase.

In general, for most clonal DNA sequencing systems, it is desirable to have as much DNA as possible on a surface, in order to maximize the amount of data signal. However, as the DNA is randomly placed on the surface, the spacing of the DNA may cause steric hindrance in a subsequent polymerization reaction.

In many different sequencing applications, target DNA or primers may be bound to the surface of the substrate. As a result of the attachment methods, the target DNA and primers may be randomly placed on the surface, and may be in sufficiently close proximity that steric hindrance occurs during a polymerase extension. In some embodiments, primers are attached, bound, or associated with the substrate while hybridized to DNA that overlaps the primer, which may provide a priming site for a polymerase. Said primer and overlapping DNA may further comprise a polymerase, which may serve as a spacer to prevent binding of said primers such that steric hindrance would occur, as for example when there is insufficient room for polymerase on each strand of DNA. Said polymerase and overlapping DNA may subsequently be removed so that a target DNA may hybridize to said substrate attached, bound or associated primer for primer extension, which may be for amplification purposes or for sequencing purposes. Alternately the primer could be in the form of a hairpin with an extended end where the polymerase could bind eliminating the need for the longer DNA. Even if a Biotin Streptavidin binding of the template is used, the size of the streptavidin (3nm) may be insufficient to properly space the DNA molecules such that there is room for the polymerase (7 to lOnm). In one embodiment, the target DNA may be appropriately spaced such that steric hindrance cannot occur. This may be achieved by for example, using a complex with double stranded DNA and a polymerase which is smaller, similar to or larger in size than the polymerase which will be used for the sequencing reaction. Alternatively, other proteins which bind to double stranded or single stranded DNA and may be of appropriate size may be used, including a polymerase that is intended for a subsequent sequencing reaction. The polymerase may be processive, so that it remains bound during the attachment process, and may further have additional binding moieties associated therewith, to further enhance the ability of the polymerase to remain in place during the DNA binding process. Alternatively, moieties other than proteins can used to space out the DNA and then removed resulting in DNA spaced to avoid steric hindrance. In some embodiments it may be desired that the surface of a substrate may not be saturated with DNA. A dsDNA strand is 20 Angstroms in diameter (2nm), as opposed to a diameter of potentially greater than lOOAngstroms for the polymerase. For example E. coli 22S RNA polymerase is 135 Angstroms in diameter (Kitano et al J. Biochemistry 1969 65 1-16); thus DNA spaced using said polymerase may be spaced at a 2 percent (2^Λ2/13.5^Λ2*100) saturation level 70

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 relative to dsDNA bound to said substrate in a saturated configuration. Polymerases do vary noticeably in size, for example, B. strearothermophilus DNA polymerase I diameter was reported to be 9nm (Kiefer et al Nature Vol. 391 15 304) in contrast with previously cited 13.5nm for E. coli 22S RNAP. In some embodiments, it may be desirable to match the number of template copies to the size of the sensor; thus in some embodiments it may be desirable to use a relatively small number of template copies. In other embodiments, it is desirable to use large numbers of template copies. Thus in some embodiments, it may be desirable to use 1,000 to 1,000,000 template copies, or 10,000 to 100,000 template copies, or 100,000 to 1,000,000 template copies. In other embodiments, it may be desirable to use 1,000,000 to 100,000,000 template copies, such as between 1,000,000 to 5,000,000 template copies, 5,000,000 to 20,000,000 template copies, or 20,000,000 to 100,000,000 template copies.

Steric hindrance may still occur when the spacing is equal to the size of the polymerase when the desired length of the DNA template is long, such that much if not most of said DNA is not perpendicular to the substrate. In some embodiments, the DNA template is sufficiently long to “ball up”, potentially taking up more space on the substrate than the polymerase. In some cases a ssDNA and/or dsDNA binding moiety may be used to space DNA, where said DNA may be a DNA primer, and where said ssDNA and/or dsDNA binding moiety may be bound to other spacing moieties whereby steric hindrance is minimized by the spacing and subsequent removal of said ssDNA and/or dsDNA binding moiety and associated additional spacing moieties. Said spacing moieties may be other proteins, or may be DNA of similar length to that desired for said DNA template. In some embodiments, which may depend on the size of the desired polymerase and the desired length of template, the level of saturation of attached DNA relative to saturated attached DNA may be 0.001 to 40%, or may be 0.01 to 15%, or may be 0.1 to 5%, or may be 0.5 to 2%.

In some embodiments, by using a polymerase or other spacer while attaching, binding or associating DNA to a substrate, particularly a polymerase that is larger than the polymerase intended to extend said substrate attached primer, steric hindrance will be significantly mitigated. In further embodiments, the polymerase may also be provided with any other helper proteins or other moieties which may serve to increase the effective size of the complex.

In some embodiments the Debye length of the read reagent solution is similar to that of deionized water, which by definition has H” and OH' concentrations of IO'⁷molar, and a resultant Debye length of 680 nm. In other embodiments, the ionic concentration of the read solution is about one micro molar, which results in a Debye length of about 300nm. This may enable reading during a reaction. In other embodiments, the read solution comprises ionic solvents which may be not wholly aqueous solvents, permitting lower charge levels in solution, thus enabling a longer Debye length, and permitting the nanobridge (described herein) to sense more of a bead. For example, the read solution may permit sensing of a bead which has a diameter greater than one micron, such as two, three, five or more microns. A read solution which comprises nonaqueous solvents may have a conductivity lower than that of distilled water, permitting a higher proportion of current to pass through the counter ions associated with DNA as opposed to current which passes through the bulk solution. The solvents may be miscible with water, and may have solubility of desired moieties; some representative solvents include DMSO, alcohols, and ethers. Said miscible solvents may have lower inherent ionic concentration, having a lower concentration of H⁺ and OH' than water. Said solvent may be used in conjunction with water, in part to provide a low concentration of hydrogen ions. In some embodiments, the ionic concentrations may be less than or equal to one micro molar, such as one micro molar to 0.5 micromolar, 0.1 micromolar to 0.5 micromolar, 0.01 micromolar to 0.1 micromolar, 0.001 micromolar to 0.01 micromolar, or less than 0.001 micromolar. In other embodiments, the ionic concentration of the read reagents may be higher than one micro molar, such as for example, 1 milimolar, 2 milimolar, 5 milimolar, or a higher concentration of ions. In some embodiments, the sensor may be able to detect changes in local charge, local conductivity, or local hydrogen ion concentration.

Many commercial buffers used for polymerization contain large amounts of Sodium or Potassium Chloride, which is not required for polymerization, and may further be heavily buffered. For example, the NEB Isothermal Amplification Buffer (IX), which is generally described as being applicable for Bst polymerase, contains 20mM Tris-HCl, lOmM (NH₄)₂SO₄, 50mM KC1, 2mM MgSO₄ and 0.1% Tween-20; NEB Phi29 DNA Polymerase Reaction Buffer (IX) contains 50mM Tris-HCl, lOmM (NH₄)₂SO₄, lOmM MgCl₂, and 4mM Dithiotheitrol. Buffering reagents interfere with pH measurement when using NanoBridges or ISFETs as sensors, and the high ion concentrations create a high background level that may interfere with measurements when using the NanoNeedle sensor.

Thus in some embodiments it is desirable to use buffer reagents with lower concentrations of pH buffers and or lower total ion concentrations.

In some embodiments, it is desirable to use very low ionic strength reagents, in order to maximize the Debye length. For such embodiments, it may be desirable to use reagents that have no more salt than is needed for the enzymatic reaction. For such reagents, it is desirable to minimize the amount of salt, for example reducing or minimizing the amount of NaCl or KC1, and using sufficient Mg. Sufficient Mg may include a concentration equal to the concentration of nucleotides used in the reagent, with additional Mg to act as counter ions for the DNA and additional Mg for Trademark*

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 polymerase in the flow cell associated with the DNA. Thus the concentration needed will be a function of nucleotide concentration, the amount and length of DNA in the flow cell, the number of polymerase molecules, and the volume of reagent used.

In some embodiments where the ionic concentration is very low, the pH may be influenced by the surrounding air. Any residual CO2 which may remain after any efforts to minimize the presence of CO2 forming carbonic acid, reduces the pH. Buffering reagents contribute to the ionic concentration; so minimizing the amount of buffering is also desired. Mitigating conflicting needs between having sufficient buffering and having sufficiently low ionic strength may be accomplished by several embodiments. One embodiment uses two buffers together, for example, combined Tris and HEPES as opposed to Tris HC1, whereby both Tris and HEPES can contribute to buffering. Ideally both buffers would have a high molecular weight/charge for reduced mobility. In another embodiment, organic reagents which may be miscible with water may be used, such as an alcohol (e.g., ethanol).

In some embodiments it is desirable to eliminate any monovalent cations such as Na+ or K+ from the buffer to avoid competitive reactions relative to divalent Mg++ causing changes to the counterion distribution on the DNA or beads.

In some embodiments, the charge associated with a bead may diminish the range of an electrical chargc/conductivity sensor, or degrade the signal to noise of the sensor. As a result, in some embodiments, it is desired to minimize the amount of charge present on the surface of the bead, by for instance changing the amount of sulfate or other negative charged moieties on the surface. In some embodiments, it may be desired to have a small amount of negative charge, so that DNA or nucleotides do not bind to the surface of the beads, but not sufficient charge so that there is not significant reduction in the dynamic range of the sensor. In other embodiments the beads may have a slight positive charge such that when DNA primers are attached the beads become negatively charged. Solutions containing solvents such as ethanol can be used to solvate the bead to allow attachment of the DNA.

In a yet further embodiment, ligation may be used rather than polymerization. Four pools of probe oligos may be used, where the first base of each probe in a single probe pool is the same. The probes may use a reversibly terminated tail, or may have a native tail, such that multiple ligations may occur, with concomitant increases in signal levels. In a manner similar to the use of multiple dNTPs and polymerase, more than one pool of oligos (with all probes starting with a single base) may be combined, again with concomitant increase in the number of ligations and signal levels. The second strand may be removed and a new primer introduced wherein the length of said primer may be shorter or longer than the length of the previous primers.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In yet another embodiment, the attached DNA molecule may have a hairpin primer, where a portion of the hairpin primer has a restriction site. Subsequently, after completion of the primer extension and associated determination of the sample DNA sequence, the restriction site may be cleaved by an appropriate endonuclease enzyme or nicking enzyme, and the extended primer may be melted off by changing one of the temperature or pH of the solution in which the sample DNA is solvated. The sample may then be re-sequenced after restoring the temperature or pH of the solution in which the sample DNA is solvated to the conditions appropriate for primer extension, including appropriate concentrations of nucleotides and cations. In an alternative embodiment, a strand displacing enzyme, or an enzyme with 5’ to 3’ exonuclease activity may be used, obviating the need to remove the second strand.

In a further embodiment, a linkage may be provided which may be chemically cleaved, obviating the need for enzymatic cleavage.

In some embodiments, it is desirable to minimize the number of counter ions associated with the polymerase and or any other helper proteins. Thus it may be desirable to substitute charged amino acids in the polymerase such as Glu, Asp, Lys, His, and Arg with vciy conservative substitution such as respectively Glu to Gin, Glu to His, Asp to Asn, Arg to Gin, His to Tyr, Lys to Arg, Lys to Gin, Lys to Glu, or with a conservative substitution such as Glu to Arg, Glu to Asn, Glu to His, Glu to Ser, Asp to Gin, Asp to Ser, Arg to Asn, Arg to Glu, Arg to His, His to Arg, His to Gin, His to Glu, lys to Asn, Lys to Ser, using the more divergent BLOSUM45 alignment. Said substitutions may be from a charged amino acid to an uncharged amino acid, or may be from an uncharged amino acid to a charged amino acid, where the amino acid which is changed from an uncharged amino acid to a charged amino acid may be adjacent to a charged amino acid of the opposite charge, whereby the charge may be shared between said charged amino acids, obviating or reducing the need for a counter ion.

In some embodiments it may be desirable to make said substitutions on portions of the protein which interact directly with the fluidic environment, as opposed to interacting with ssDNA to which said protein may be bound. In some embodiments it may be desirable to add additional positively charged amino acids in locations which interact and bind with said ssDNA in order to provide tighter binding.

For example as seen in Figure 10 (Hollis et al PNAS 98 17 9557), which shows in darker grey as part of 1002 the positively charged portions of a single SSB (normally ternary in vivo), which thereby shows the portions of said SSB which bind ssDNA. Figure 10 conversely shows in darker grey as part of 1000 the portions of said SSB which may be positively charged, and which may therefore cause additional counter ions to accumulate as the SSB binds to the ssDNA, causing an increased background current change due to either the influence of charge on the sensitive area of a

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

NanoBridge, or by locally increasing the conductivity of the volume proximate the ssDNA. In some embodiments, a DNA binding protein, which may be a polymerase, or a protein with ssDNA or dsDNA binding affinity, may be mutated such that the binding of said protcin(s) to a DNA strand(s) of interest results in a lower background current change than might occur if a native DNA binding protein(s) were to bind to said DNA strand(s) of interest. In some embodiments, the background current change occurs as a result of the binding of a mutated protein(s) to said DNA strand(s) of interest may result in no observable change in the background current relative to the background current without a protein(s) being bound to said DNA strand(s) of interest as a result of the reduced charge of the mutated protein which interacts with the fluid, or as a result of the mutated protein(s) reducing the number of counter ions which may be associated with said DNA strand(s) of interest as a result of increased charge interaction between said mutated protcin(s) and said DNA strand(s) of interest. In other embodiments, the background current change occurs as a result of the binding of a mutated protein(s) to said DNA strand(s) of interest may result in a decrease in the background current relative to the background current without a protein(s) being bound to said DNA strand(s) of interest as a result of the reduced charge of the mutated protein which interacts with the fluid, or as a result of the mutated protein(s) reducing the number of counter ions which may be associated with said DNA strand(s) of interest as a result of increased charge interaction between said mutated protein(s) and said DNA strand(s) of interest.

In other embodiments, a primer may be provided which has a nick site. In still further embodiments, multiple adjacent primers may be provided, obviating the need for a nicking endonuclease. The primers may be complementary to a ligated primer, or may be complementary to a targeted section of DNA. The sequencing primers may comprise all or part of primers used for clonal generation via an amplification reaction, or may comprise regions which may be not used as part of the primers for amplification, or may comprise both regions used for primers in an amplification reaction, and a region which is not used for amplification reaction.

In some embodiments, the quality of measurements the signal may be dependent on the sharpness of the reagent front hitting the sensor, and the collection time window size may be dependent on when the reagent contacts the sensor, particularly for detecting a transient pH change as a result of a base incorporation. The sharpness of the wavefront may be reduced, by for example, diffusion while the reagent is moved through fluidic lines, valves and the flow cell itself. To better control the timing and to minimize the diffusion, particularly to minimize the time needed to effectualize a reagent delivery an electric field repulsive charge may be used to hold the active reagents, for example the dNTPs may be held away from the sensor in the axis perpendicular to the plane of the sensor array during the initial delivery along the length 75

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 of the flow cell from the input port to the output port. Later the electric field can be turned off or reversed to draw the reagents to the sensor. In some embodiments the electric field can be activated on different sections of the sensor array at different times to allow better control of the readout time window.

In some embodiments, the flow of reagents through the flow cell may be laminar, and rcagcnt(s) delivered to/on the top of the reaction chamber will stay there with mixing occurring only due to diffusion in the axis perpendicular to the flow of the reagent. This may be used to retard incorporation until desired. In some embodiments both laminar and electric fields could be combined to control the delivery to the sensor.

In other embodiments the temperature can be kept low, such that polymerase activity is minimized while the reagents are delivered and then brought up to allow' the reaction to start.

To minimize diffusion effects, all nucleotides needed for sequence analysis may be present in the system, and the only reaction triggering factor may be magnesium ions. Magnesium ions have a higher diffusion rate than other components of the reaction such as polymerase and dNTPs. In these or other embodiments, the temperature of the reservoir and reaction may be kept below the activation temperature of the polymerase enzyme. Reactions can thereby be triggered by precise temperature control, thereby overcoming diffusion limitations. For example, in these embodiments the parallel reactions of the array may be almost simultaneous. Activation temperatures are known in the art and/or may be determined experimentally for any particular embodiment. The activation temperature for Klenow polymerase is approximately 4° C, and the activation temperature for Taq is approximately 60° C. In still other embodiments, the reaction may be triggered by introduction of a required reaction cofactor, which may be sequestered in nanoparticles or vesicles prior to the reaction, and released with the appropriate external stimulus (e.g., laser or temperature).

In some embodiments, as a part of the sample preparation process, “barcodes” may be associated with each sample. In this process, short oligos are added to primers, where each different sample uses a different oligo in addition to primer. The primers and barcodes are ligated to each sample as part of the library generation process. Thus during the amplification process associated with generating each colony, the primer and the short oligo are also amplified. As the association of the barcode is done as part of the library preparation process, it is possible to use more than one library, and thus more than one sample, in generating the clonal populations. Synthetic DNA barcodes may be included as part of the primer, where a different synthetic DNA barcode may be used for each library. In some embodiments, different libraries may be mixed as they are introduced to a flow cell, and the identity of each sample may be determined as part of the sequencing process.

Sample separation methods can be used in conjunction with sample identifiers. For example a chip could have 4 separate channels and use 4 different barcodes to allow the simultaneous running of 16 different samples. This permits the use of shorter barcodes while still providing unambiguous sample identification.

Nanosensors and Detection Methods and Systems

In some embodiments, a charge or pH sensitive detector is used to determine the sequence of a DNA colony. A colony may be generated on a bead, the bead may be transferred to a sensor location, provided with primers, polymerase, and dNTPs while observing the change in charge or pH due to the incorporation of dNTPs. There may be a one to one correspondence between a sensor location and a colony.

In embodiments of the devices disclosed herein, a plurality of nanonccdlc sensors arc employed having at least one electrode formed in the shape of an arc conforming to the edge of a depression where one of the plurality of magnetic beads sits.

The nanosensor is a sensor designed to detect beads or particles less than one of 0.1, 1,5, 10 or 20 micrometers as measured on the diameter or the long axis for nonspherical beads or particles. Alternatively, the sensor may be sensitive to moieties associated with said beads or particles, or with reaction products or byproducts wherein the reaction includes a moiety associated with said bead or particle. Said moieties may include DNA fragments, hydrogen ions, or other ions which may be counter ions and thus associated with said beads or particles or moieties bound or associated with said beads or particles. Nanosensors can include NanoBridge, NanoNeedle or ISFET sensors. A NanoNeedle may be an impedance measuring sensor including two electrodes situated to measure the conductivity of the local environment between the active area of the electrodes. “NanoBridge” refers to a resistive device which may include a sensor which may respond to charge proximate to the active area of said resistive device, and wherein said resistive device may further be a semiconductor device.

In some embodiments, the NanoNeedle functions as a pH sensor, as described in US Provisional Application 61/389,490 titled “Integrated system and methods for polynucleotide extraction amplification and sequencing.

The sensors may be used for detection of transient properties associated with incorporation events as described in US7,932,034, which is hereby incorporated by reference in its entirety.

In embodiments of the devices disclosed herein, a plurality of nanobridge sensors are employed having an active area partially encircling, and in immediate 77

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 proximity to, one of the plurality of magnetic beads. For example, the radius of the active area maybe less than the radius of the magnetic bead.

In embodiments, of the device, the plurality of nanobridge sensors are adapted to measure the incorporation of nucleotides into a polynucleotide, and the nanobridge sensors each have an active area and a conducting element. The conducting element having a work function that matches a work function of the active area.

In some embodiments of the current invention, the sensors may be NanoNeedle sensors, nanobridge sensors, ISFET sensors, ChemFETs, nanowire FETs, carbon nanotube FETs, other types of charge, conductivity, or pH detecting sensors, or a combination of different types of sensors at each sensor location wherein a bead or colony may be located for sequencing reactions. An individual sensor may detect using only a single modality such as charge, conductivity, or pH, or an individual sensor may detect using more than one modality, such as responding to both charge and pH. The sensors may provide similar information, or they provide complementary information. For example, one sensor at a sensor location may respond to changes in pH, while another sensor at a sensor location may respond to changes in conductivity. In some embodiments, one sensor may detect local changes in pH, conductivity, or charge, while another sensor is used as a reference. In one embodiment, the reference sensor may be placed so that it does not contact any reagents, and may be used to compensate for changes in temperature, power supply voltages, etcetera.

The change in charge concentration may result from other sources, including binding of DNA to DNA attached directly to the sensor, which may be either a nanobridge, a NanoNeedle, or a FET, or may result from binding of charged cDNA, RNA, proteins, lipids, carbohydrates; the change in charge may also result from an enzymatic reaction, or any other chemical reaction which may be sufficiently localized as to largely occur within the sensing region of one sensor, and within the sensing region of another sensor.

In some embodiments, a combination of NanoNeedle sensors, nanobridge sensors, and magnetic retention structures are used. The NanoNeedle structures may be located under the magnetic structures that make up the magnetic array elements, or the NanoNeedle structures may be located on top of the magnetic structures that make up the magnetic array elements. In other embodiments, the NanoNeedles may be located orthogonally, or at some other angle with respect to the structures that make up the magnetic array elements.

In some embodiments the NanoNeedle sensor may measure the impedance change due to ions generated by a DNA polymerization event.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In other embodiments, the NanoNeedle measures the impedance surface change due to the incorporation of the DNA. Each base of DNA has a negative charge. As bases are added the charge becomes more negative. This additional charge attracts positive counter-ions which can change the conductivity of the surface of the DNA coated bead. This impedance change may also result from molecules that bind to DNA. Because the charge is associated with a fixed molecule (DNA bound to bead) the local fluid environment is changed from the polymerization condition. For example, one buffer can be used for the base incorporation and a second buffer can be used to measure the conductivity change from a previous measurement.

in some embodiments a NanoNccdlc sensor is configured to measure the impedance change of a bead as bases are added to the template DNA attached to said bead. In other embodiments, said DNA template may be attached or associated with the substrate or a coating on said substrate, or may be attached or associated with toe device electrodes or coatings on said electrodes. To improve performance it is desirable to reduce the other impedances.

The sensor impedance may be dominated by other impedances. For example, the impedance of the bulk reagent between the electrodes and the Debye layer associated with the bead may be large relative to the impedance through the Debye layer associated with the bead if the physical alignment is not good between the electrodes of a NanoNeedle and a bead to which DNA is bound. If, for example, the impedance of the bulk reagent constitutes 90% of the total impedance between the electrodes, and the impedance of the DNA on the bead and its associated counter ions constitutes 10% of the total impedance between the electrodes, a 1% change in the impedance of the DNA and associated counter ions will result in a 0.1% change in toe total impedance between the electrodes. The impedance of the bulk reagent between the electrodes may be small relative to the impedance through the Debye layer associated with the bead if the ion concentration of the bulk solution is high.

Figure 16 schematically illustrates a simplified circuit 1600 for a NanoNeedle sensor with bead. The sensor may have parasitic capacitance 1614 and parasitic resistance (not shown). The sensor may further have double layer capacitances 1610 associated with each electrode 1612. The resistance resulting from the counter ions associated with the DNA or other sample bound, attached or associated with the bead 1606, and may be in parallel with the resistance resulting from the bulk fluid 1602, the resistance from conductivity through the counter ions associated with charge on the surface of the bead 1606, and the resistance from conductivity through the counter ions associated with charge on the surface of the sensor 1608. Additional portions of the circuit that may add complexity include resistances (not shown) between the double layer capacitance 1610 and the resistance from conductivity through the counter ions

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 associated with charge on the surface of the bead 1606, and the resistance from conductivity through the counter ions associated with charge on the surface of the sensor 1608.

Thus in some embodiments, it is desirable to minimize the distance between both electrodes and the bead. In other embodiments, it is desirable to measure the counter ions from as much of the bead as possible, allowing averaging from as much of the entire surface of said bead as possible. In some embodiments it may be desirable to position said electrodes at opposite sides of said bead, allowing current to flow over the entire surface of said bead in as even a manner as possible. In other embodiments, it is desirable to have electrodes that arc not as small as might be possible, so that the current density of the current path emitting from said electrodes is not significantly higher than the current path at the point in the current path wherein the current density is smallest. For example, if an electrode could be made infinitely small, the current density emanating from said infinitely small electrode would be infinitely large. Tn another example, the electrodes comprise spherical caps at opposite ends of said bead, and the circumference of the circle formed by the spherical cap is one half of the length of the great circle of said bead. The maximum current density will be twice as high at the spherical cap as it is at the great circle midway between the spherical caps. In some embodiments the ratio of maximum current to minimum current over the surface of the bead may be 2 to 1; in other embodiments, the ratio of maximum current to minimum current over the surface of the bead may be between 2 to 1 and 3 to 1, between 3 to 1 and 4 to 1, between 4 to 1 and 6 to 1, between 6 to 1 and 9 to 1, between 9 to 1 and 15 to 1,between 15to 1 and 30 to 1, or between 30 to 1 and 100 to 1.

In some embodiments, the electrodes are fabricated with semiconductor technologies, and the area of the electrode adjacent to the bead is of a height equal to the thickness of the electrode. It may be desirable to hold the electrode a small distance from the bead, such as from 0.1 Debye lengths, to 0.3 Debye lengths, from 0.3 Debye lengths, to 1.0 Debye lengths, from 1.0 Debye lengths, to 3.0 Debye lengths, from 3.0 Debye lengths, to 10.0 Debye lengths, or from 10.0 Debye lengths, to 100 Debye lengths. The Debye length is considered to be an additive combination of the Debye length of said bead and said electrode. Alternatively, it may be desirable for the electrode to have a length that is a fraction of half the circumference of the great circle of the bead. Said fraction may be from 0.01 to 0.03 of half the circumference of the great circle of said bead, from 0.03 to 0.1 of half the circumference of the great circle of said bead, from 0.1 to 0.3 of half the circumference of the great circle of said bead, from 0.3 to 0.75 of half the circumference of the great circle of said bead. In some embodiments, the system maintains a distance between the electrode and the bead, and fabricates the electrode so that it is a fraction of half the circumference of the great circle of the bead.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments, it may be desirable to reduce the current passing through the bulk reagent, in order to maximize the portion of the measurement current which passes through the counter ions associated with DNA bound to, attached to, or associated with a bead. As a result, in some embodiments, it is desirable to physically reduce the volume of bulk reagent in proximity to the bead, such that the impedance contribution from the DNA counter ions is maximized. In other embodiments, it is desirable to minimize the surface area of the structure which retains the bead in proximity to the electrodes for measurement of the DNA counter ions. In a further embodiment, it is desirable to minimize the zeta potential of the bead and or surface(s) of the structure which retains the bead in proximity to the electrodes for measurement of the DNA counter ions.

The NanoNeedle structures may be fabricated in an array of NanoNeedles, permitting large numbers of single DNA molecules or colonies to be sequenced at the same time.

As shown in Figure 17, a NanoNeedle sensor structure 1700 may be fabricated with a Silicon substrate 1701, and may have a 800 nm deep channel 1702 etched in said substrate. A silicon oxide layer of 200nm thickness 1703 may be fabricated on the substrate, followed by a conductive p+ silicon layer of 80nm thickness 1704, followed by a silicon oxide layer of 30nm thickness 1705, followed by a conductive p+ silicon layer of 80nm thickness 1706, followed by a silicon oxide layer of 20nm thickness 1707. The channel may be created after the structure is fabricated. The structure may be generated such that an oxide layer or a resist layer covers all sections which may be to be retained in the final structure. A chemical wet etch, a plasma etch, or a vapor phase etch may be utilized to remove the silicon or other similar substrate from under the structure. The conductive tip of the structure may then be exposed using an ion milling step.

All of the thicknesses may be varied, as may the materials. The channel in the substrate may alternatively be fabricated using an oxide layer, with a resist layer in the volume of the channel. The layers of Oxide and conductors may then be fabricated on top of the oxide and resist, obviating the need to under-etch the structure. Figure 18 illustrates a single ended NanoNeedle array fabricated in a manner similar to that schematically depicted in Fig. 17.

As shown in Fig. 19 such a structure may have sensors 1901 on both sides of a channel 1902 formed in a substrate 1903. Polymerase and or target DNA 1904 may be attached to the active area of the sensor. The sensor itself may be used to electrophoretically and or dielectrophoretically localize the polymerase and or target DNA to the active area of the sensor. The target DNA may be a single double stranded, single stranded DNA target, or a circularized DNA target, or a local amplification may 81

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 be done in place on the active area of the sensor, as described in PCT/US2011/054769. Fig. 20 illustrates an interdigitated NanoNeedle array fabricated in a manner similar to that schematically depicted in Fig. 19.

Nucleotides or probes 1905 may be then be provided, and a sequencing by synthesis process, or a sequencing by ligation process may commence.

To improve the sensitivity of either the NanoNeedle or the NanoBridge, a local amplifier may be provided. The amplifier may be either a BJT or an FET. In some embodiments, an amplifier is used with one amplifier circuit for each sensor or with multiple sensors sharing the same amplifier. In other embodiments, some amplification may be associated with each sensor, and additional amplification and or other associated circuitry is shared or multiplexed between different sensors. The sensor can be fabricated as a narrow structure, and can be etched under the structure so that both sides may be accessible to changes in pH, or to changes in conductivity. The surface of the device may be rough, permitting greater surface area for binding of sample molecules. The surface associated with the electrodes of a NanoNeedle may be gold or platinum, or may be platinum black, iridium oxide, or Ppy/PSS to increase the surface area and the associated double layer capacitance.

Electric concentration of ions may be effected, concentrating the DNA, polymerase, primers nucleotides and other reagents as needed to the active area of the NanoNeedle or NanoBridge sensor. Said concentration allows much more of the sample to be attached or associated with each sensor, mitigating the need for whole genome amplification.

Another factor which may prevent optimal measurement of the impedance of DNA on the bead includes counter ions which result from the Debye layer associated with the zeta potential of the surface of the bead and or the counter ions which result from the Debye layer associated with the zeta potential of the surface of the sensor. These counter ions may result in a current which may be in series and or in parallel with the desired current associated with counter ion of the DNA on the bead. Further, as the zeta potential varies, the Debye length and the number of associated counter ions may vary concordantly. Said zeta potential may vary with changes in buffer conditions including changes in pH, salt concentration and various other factors. Said concordant variation may thus confuse measurements of the DNA. Thus it is desirable to both minimize the zeta potential, and to minimize variations of zeta potential with variations in buffer conditions.

In some embodiments, the sensor may be fabricated with silicon. Silicon dioxide has a significant zeta potential magnitude at pHs typically useful for polymerization activity, such as pH 7 to 9; but the zeta potential magnitude of silicon

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 nitride is significantly less than that of silicon dioxide. Thus in some embodiments, it silicon nitride is used at the interface between the silicon sensor device and any components which may come into contact with the silicon sensor device, to thereby minimize the zeta potential and concomitant current through counter ions which may be in the associated Debye layer.

In some embodiments, a coating is applied over the surface of the sensor. The sensor may be fabricated of silicon, silicon dioxide, PDMS, Topaz™ or other various polymers or combinations thereof, where said electrode and or coating may comprise materials such as TiO², ZrO², or Indium Tin Oxide, BaTiO, such that the zeta potential and the resulting Debye layer arc significantly reduced. In other embodiments, surface coating(s) such as PEG (Poly Ethylene Glycol), PTFE, poly L lycine, acrylate, methyl cellulose, n-dodecyl-B-D-maltoside, acrylamide, fluorinated alkane chains, or other cross-linked or partially cross-linked polymers are incorporated to modify the zeta potential, or combinations of surface coatings are used to similarly minimize the zeta potential and concomitant Debye length. In other embodiments the zeta potential magnitude is reduced by protecting the silanol groups with a compound such as trimethylchlorosilane which decreases the number of ionizable silanol groups.

In some embodiments, it is desirable to reduce the zeta potential of a bead on which DNA to be sensed is attached, thereby reducing the concomitant current resulting from counter ions associated with the surface of the bead due to said zeta potential. Thus in some embodiments it is desirable to fabricate the bead of a material with a low zeta potential at the pH levels anticipated for effective polymerization, or the bead may be coated with a material with a low zeta potential at the pH levels anticipated for effective polymerization, such as such as PEG (Poly Ethylene Glycol), PTFE, poly L lysine, acrylate, methyl cellulose, n-dodecyl-B-D-maltoside, acrylamide, fluorinated alkane chains, or other cross-linked or partially cross-linked polymers are used to modify the zeta potential, or combinations of surface coatings may be used to similarly minimize the zeta potential and concomitant Debye length.

In some embodiments, it is desirable to minimize variations in pH which may result from buffer reagents, while it may be simultaneously desirable to minimize ionic concentration. As a result, it is desirable to use reagents with little buffering capacity while maintaining a fixed pH. Buffers may in some cases be degassed as part of an assay or method; said buffers may then be subject to changes in pH as CO2 dissolves into the buffer reagent. In some embodiments, it may be desirable to restrict the interaction between CO2 and buffer reagents. Thus it may be desirable to exclude atmospheric gases, and to provide other gases which do not include CO2 such as Nitrogen, Argon, or other purified gases, or mixtures of gases which do not contain CO2

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 or other gases which might otherwise dissolve into said reagent buffer, and thus change the ion concentration and or pH.

In some embodiments, the system includes an external gas source such as an industrial gas cylinder. In some embodiments said gas cylinder is external to an instrument where the fluidics resides. In other embodiments, the industrial gas cylinder is placed within a compartment within the instrument. In other embodiments, a CO²scrubber/degasser/debubbler is used such as a regenerable metal oxide system, a Kraft process system, an activated carbon system, a membrane system which may use a membrane such as the Systec AF® or Poridex™. Said CO? scrubbcr/dcgasscr/dcbubblcr may be built within said instrument, or may be external to said instrument.

In some embodiments, it may be desirable to bring the sensor electrodes for a sensor such as a NanoNeedle sensor into close proximity to a bead, in order to minimize the amount of bulk reagent volume between the NanoNeedle electrodes and the bead. One embodiment may have a bead held in a depression as shown in Figures 2A, 2B and 2C. The depression may be formed of a material which is deposited on a substrate, and the material forming the depression may have a pair of NanoNccdlc electrodes formed upon said material. The electrodes may be formed in an arc conforming to the edge of the depression, and thus to the edge of the bead. In some embodiments said depression may be fully accessible to fluids on one side, or two sides, or said depression has a width to depth ration of less than 1.0, or may have a width to depth ratio of 1.0 to 0.9, 0.9 to 0.8, 0.8 to 0.7, 0.7 to 0.6,0.6 to 0.5, 0.5 to 0.4,0.4 to 0.3,0.3 to 0.2,0.2 to 0.1 or 0.1 to 0.01.

In some embodiments, the depression minimizes the volume of reagent in proximity to the bead. The depression may be shaped so as to conform to the shape of the bead, whereby the bottom of the depression is narrower than the cross section of the depression at the height of the electrodes. In other embodiments, the electrodes are covered by a layer of additional material, such that the effective depth of the depression is greater than half the diameter of the bead, further reducing the volume of bulk reagent proximate the bead.

The electrodes may thus be touching the surface of the bead, or may be within the Debye length of the surface of the bead or particle and the DNA attached or bound thereto. In some embodiments, the electrodes are curved such that the electrodes follow a curve with a radius similar to the radius of the bead, permitting better coupling between the electrode and the bead. Such a device permits a minimum influence on the total impedance between the NanoNeedle electrodes by the bulk reagent solution, and a maximum influence by the DNA attached to or bound to the surface of the bead or particle or the counter ions near the DNA.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments, a NanoNeedle has the active area of the sensor shaped to fit a bead or other sample retaining mechanism. It may, for example be shaped in an arc, with the curve of the arc oriented so as to align with the curve of a bead. It may also have one NanoNccdlc of a NanoNccdlc pair configured such that it is offset or “shorter” than the other NanoNeedle of said NanoNeedle pair, such that the inner radius of the arc has a larger diameter, and the same centroid. Said offset may permit an increase in the volume of the sensing region associated with the NanoNeedle pair, and may further change the orientation of the field associated with the sensing region and thus orientation of the sensing region, so that the sensing region is more oriented towards the center of the bead, rather than parallel to substrate.

In an alternative embodiment as shown in schematic side viewFigure 21A and schematic top view Figure 21B, one electrode 2105 may be attached directly to the substrate 2102 or on another layer upon said substrate, allowing isolation from said substrate. The second electrode 2105 fh the NanoNeedle may be attached upon a dielectric 2113 portion of the sensor which is utilized to position the bead or particle 2101 in a fixed location. The bead or particle 2101 is thus in contact with both electrodes 2101,2105, minimizing the influence of the bulk reagent solution on the total impedance between the NanoNeedle electrodes, as opposed to the impedance resulting from the counter ions within the Debye length associated with the bead or particle and the DNA which is attached or bound to the bead or particle.

In some embodiments, one or both electrodes may be fabricated such that said electrodes conform to the curve of the bead so as to provide a lower and more regular impedance between the electrode(s) and the bead. The curvature may abut the edge of a depression, or may be slightly farther from the edge of the depression so as to allow a larger area of interface and a lower current density.

In a further embodiment as shown in Figure 21C, the bead or particle 2101 may be held in place on a substrate 2102. A first electrode 2105A of a NanoNeedle 2100 may be attached directly to the substrate 2102, or to an adhesion layer (not shown) adhered to said substrate 2102. A dielectric layer 2114 may then be fabricated so as to cover said first electrode 2105A. A second electrode 2105B of a NanoNeedle 2100 may then be fabricated above the dielectric 2114 and said first electrode 2105A of the NanoNccdlc 2100. The second electrode 2105B may be shorter, so as to conform to the curve of the bead or particle. The difference in the length will be a function of the diameter of the bead or particle 2101, and the thickness of the two electrodes 2105 and the dielectric 2114 between the electrodes 2105. In this manner the electrodes 2105 may be in contact with the bead or particle 2101, or may be in very close proximity to the bead or particle 2101, such that the impedance resulting from the counter ions within the Debye length associated with the bead or particle 2101 and the DNA which

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 is attached or bound to the bead or particle 2101 is greater than the impedance of the bulk reagent.

In a further embodiment, the electrodes are fabricated such that said electrodes do not abut the edge of a depression, but may be instead fabricated a short distance from the edge such that the current density in immediate proximity to the electrode may be reduced.

In Figure 22A, a NanoNeedle 2200 is schematically illustrated in a side view wherein said NanoNeedle 2200 has electrodes 2205 on each side of a depression in a dielectric 2203, wherein a bead 2201 may be retained on a substrate 2202, and metalization 2204 to said electrodes 2205 may be used. As shown in Figure 22A, the dielectric material 2203 may be similar in thickness to one half the diameter of the bead 2201, and the depression width may be slightly larger than the diameter of the bead 2201 while still allowing the bead 2201 to be within the Debye length of said bead 2201 with respect to both electrodes 2205. Said thickness of the dielectric 2203 may of a thickness which permits retention of the bead 2201, and maintains said bead 2201 within a Debye length of said bead 2201 of both electrodes 2205.

In Figure 22B, a NanoNeedle 2200 is schematically illustrated in a side view where the NanoNeedle 2200 has electrodes 2205 on each side of a depression in a dielectric 2203, where a bead 2201 is retained above a substrate 2202, and metalization 2204 to the electrodes 2205 is used. As shown in Figure 22A, the dielectric material 2203 may be less than one half the diameter of the bead 2201, potentially the dielectric material 2203 thickness is one quarter to one third the diameter of the bead, and the depression width may be less than the diameter of the bead 2201 such that the bead 2201 is suspended above the substrate 2202. The close proximity of the bead 2202 and the electrodes 2205 maintains a spatial proximity between the bead 2202 and the electrodes 2205 such that the bead 2201 is within a Debye length of the bead 2201 of both electrodes 2205.

In Figure 22C, a NanoNeedle 2200 is schematically illustrated in a top view wherein the NanoNeedle 2200 has electrodes 2205, which are curved to maintain close proximity to the bead as a result of the depression in the dielectric material 2203 being smaller in diameter than the diameter of the bead 2201, such that the bead is held in immediate proximity over an arc corresponding to the point of contact or close proximity between the bead 2201 and the electrodes 2205.

Figure 22D is a three transparent dimensional drawing of a NanoNeedle structure 2200 similar to that of Figure 22B and Figure 22C, but with substantial fluidic access to the bead 2201. The bead 2201 is held suspended above the substrate 2202, and is instead held against electrodes 2205A, which are fabricated above the dielectric

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 material 2203, where the electrodes 2205A and dielectric material 2203 are shaped to be within a Debye length of the bead 2201 Debye length, but is not curved to match the curvature of the bead 2201, such that instead of having a line contact between the bead and the electrode 2205 and or dielectric material 2203, there arc three or four point contacts between the bead 2201 and the electrode or dielectric material 2203. Figure 23D also depicts magnets 2208 which apply the force to retain the bead 2201 in place in the NanoNeedle structure 2200.

A NanoNeedle may be configured to be in a double spiral or serpentine pattern, in order to increase the length, and simultaneously decrease the width of the nanobridge channel. A sensing region that is too wide will have a comparatively low impedance, and may have areas of the sensing region which have smaller changes in local charge density than other regions, for example, at the edges of a bead in comparison with the center of a bead. The sensing region which is “too wide” may thus also have smaller changes in impedance, as only a portion of the sensing region may be significantly affected by a binding or reaction which results in a local change in charge. In contrast, a NanoNeedle that is too long and thin may have an impedance that is so large that any current change may be too small to sense with good signal to noise. Thus the width and length of the channel associated with a nanobridge sensor will need to be tuned for the specific application for which said nanobridge sensor is intended.

In some embodiments, a NanoNeedle is configured to have several active regions as part of a single NanoNeedle. The active regions are located at various locations with respect to a single sample, providing an average of several different areas from the sample region, such that variations in locations of a sample region for example, slight misalignment of a bead relative to a sensor or variations in loading densities on a surface, will have less affect on the signal to noise for a sensor.

Streaming potential was originally observed by Quinke in 1859, and is a well known phenomenon in capillaries; it is a function of the flow rate, the zeta potential, and the conductivity of the fluid, among other factors. Thus a voltage may be impressed on a NanoNeedle, and variations in the flow rate or distances between electrodes may result in variations either spatially or temporally in the bias impressed on a NanoBridge.

In other embodiments it is desirable to orient NanoNeedle electrodes parallel to the flow of the fluid, so that there will not be a potentially variable streaming potential impressed between the electrodes of the NanoNccdlc, as would be the case if the electrodes were orthogonal to the flow of said fluid.

In some embodiments, the NanoNccdlc is coupled with a local capacitor, or capacitor, associated with one or both electrodes, in order to prevent influence from DC

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 bias levels from the driver circuit or leakage from within the chip sensor from influencing the output signal.

In a further alternative embodiment, a Nanobridge sensor structure 2300 as shown in Figure 23 is used instead of the aforementioned NanoNeedle sensor. The Nanobridge sensor may be used in the same manner as the NanoNeedle, including with circularized or linear DNA, a linear or hairpin primer, a polymerase as described for the nanobridge, and may fabricated as an array.

The NanoBridge sensor structure 2300 may comprise a silicon-on-insulator device, comprising a substrate 2360, a dielectric insulator 2310, two higher doped semiconductor regions 2304A and 2304B, a lower doped semiconductor active area region 2305, a further metalization layer 2340 which may cover said semiconductor active area region 2305, and which may have a further dielectric coating 2350 over said semiconductor active area region 2305.

In some embodiments, the nanobridge senses local changes in charge. Changes in surface charge of a surface of the nanobridge abutting the flow cell may result from changes in charge in the second layer in the flow cell. These changes in charge on the surface of the nanobridge may then change the charge distribution in the nanobridge, and thus change the conductivity of the nanobridge. The area of the nanobridge surface which has charge changes may thus have changes in conductance in the associated volumes of the nanobridge, while other surface areas of the nanobridge may not have changes in surface charge, and thus may not have changes in conductance in the associated volumes thereof. Depending on the type of semiconductor material the nanobridge is constructed of (n or p type), the amount and uniformity of doping in the nanobridge semiconductor material, and the sign of the charges (positive or negative) on the surface of the nanobridge, and whether the change in charge is an increase or decrease in the amount or density of surface charge may increase or decrease the conductance.

In some embodiments, a change in the charge on a bead located within the Debye length causes a corresponding change quantity or concentration of charge locally present in both of the layers of the double layer. Said change in the quantity of charge relates directly to the local ion concentration, and thus also to the surface layer capacitance, and the conductance of the reagent within the Debye length. Said change in the charge may be either an increase, or a decrease depending on the relative charge of the surface layer and the charge change on a bead.

In some embodiments for sequencing of a clonal bead, a Nanobridge sensor is used. The Nanobridge sensor may be used in a similar fashion to the NanoNccdlc.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In an alternative embodiment, the nanobridge detects a local temperature change, and thus acts in part as a temperature sensor.

In an alternative embodiment the Nanobridge can be configured to operate as a temperature sensor and/or a pH sensor to detect nucleotide incorporations. This method is further described in US patent application 20080166727 titled “Heat and pH measurement for sequencing of DNA,” which is hereby incorporated in its entirety.

The present invention provides methods and systems for polynucleotide sequencing based upon pH and/or temperature detection. In some embodiments, the system and method may further employ (or alternatively employ) dyes or quantum dots that allow visual or optical detection of pH and/or temperature changes. This monitoring may allow monitoring of the bulk solution, or may allow local monitoring of the volume associated with each colony, or may allow for monitoring of both the bulk solution and the volume associated with each colony.

In other embodiments, an array of NanoBridge sensors is etched underneath, so as to further minimize the channel size, and to maximize the surface area which interacts with the charge resulting from the DNA sequencing reaction. In other embodiments, the array of NanoBridge sensors may not be etched underneath, or may be partially etched so as to provide a more robust structure. In yet further embodiments, the array of NanoBridge sensors is configured such that it is arranged in a comb configuration, with sensors interleaved between each other from both sides, with potential features, such as a potential amplifier arranged alternatively on one side, and then the other. In another embodiment, the array of Nanobridge sensors is arranged such that potential features, such as an amplifier, are all arranged on one side of the sensor array.

In some embodiments, a nanobridge sensor is configured such that the width and length of the sensing channel is aligned for optimal sensitivity for a sensing application. Variations may relate to the spacing and size of a sample region, the charge associated with sensing the desired moiety, and the impedance of the nanobridge in nonsensing regions, such as conductive portions of the nanobridge between the sensing region and a local amplifier, or other associated impedances.

In some embodiments, the sensor is a NanoBridge sensor 2400 where the active area is fabricated such that the active area partially encircles the bead or particle 2401, and is in immediate proximity to the bead or particle 2401, as shown in Figure 24A, Figure 24B, and Figure. 24C. The sensor may comprise a substrate 2402, on which a layer of dielectric and or semiconducting material 2403 may be applied. The active area of the NanoBridge 2405 may be fabricated such that it largely encircles the bead or particle 2401. Metalization lines 2404 may connect to more highly doped regions of

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 semiconducting material 2404A which then interface with the active area 2405 of the NanoBridge. Figure 24A is a side view of a “ring” NanoBridge, where the inner portion of the active area 2405 is within the Debye length of the bead or particle and the DNA which may be bound thereto. The active area may be partly or entirely within the Debye length of the bead or particle, resulting in impedance of the entire active area changing in response to changes in the charge which is bound or associated with the bead or particle and/or the incorporation event of a nucleotide or nucleotide analog. The diameter of the ring and the associated supporting structure 2403 may be sized such that a bead fits closely within the ring.

Alternatively, as shown in Figure 24B, the ring and support structure 2400 may be sized to be smaller than the diameter of the bead or particle2401, such that a bead may rest upon the ring of the active area of the NanoBridge 2405, particularly when held by a magnetic field, insuring that the ring is within the Debye length of the bead or particle 2401 and the DNA bound thereto. Figure 24C is a top view of a NanoBridge 2400 implemented with a ring structure, showing the overlap of the bead 2401 over the active area of the NanoNeedle 2405, and the electrical conductors 2404 which provide a means to measure the impedance of the active area 2405.

In some embodiments, the shape of the Nanobridge sensors are optimized to provide greater interaction with the magnetic or paramagnetic particle. Said Nanobridge sensors may be shaped in a spiral, serpentine or other non linear shape, or a shape that has a variable cross section so as to provide more surface area while retaining a narrow channel for current to flow through in the channel of the Nanobridge.

The electrical conductors 2404 may be connected to heavily doped regions of the NanoBridge (not shown), which then provide electrical connection to the active area of the NanoBridge 2405. Alternatively, the electrical conductors 2404 of the NanoBridge may be directly connected to the active area 105 of the NanoBridge with having an Ohmic connection by fabricating the NanoBridge electrical conductors 2404 such that the work function matches the work function of the active area of the NanoBridge 2405. The value of the work function of aluminum is close to value of the work function of lightly doped silicon, but is not a perfect match. To create a more perfect match, an aluminum alloy may alternatively be utilized.

Streaming potential was originally observed by Quinke in 1859, and is a well known phenomenon in capillaries; it is a function of the flow rate, the zeta potential, and the conductivity of the fluid, amongst other factors. Thus a voltage may be impressed on a Nanobridge ISFET, or other chemFET sensors, and variations in the flow rate or distances between electrodes may result in variations either spatially or temporally in the bias impressed on a NanoBridge.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments it may be desirable to use reference electrodes which may be fabricated between different NanoBridge or ISFET sensors in an array in order to reduce the variation in the bias voltage impressed on the sensitive area of the sensor. In some embodiments, reference electrodes arc fabricated in between each NanoBridge or ISFET in an array NanoBridges or ISFETs on the substrate of the array of said NanoBridges or ISFETs. Said electrodes may be interconnected by metalization as part of the fabrication of said Nanobridge or ISFET array. In other embodiments, sets of said electrodes are interconnected using metalization as part of the fabrication of said NanoBridge or ISFET array. In other embodiments, the reference electrodes are fabricated such that a fixed or variable number of NanoBridge or ISFET sensors are between each NanoBridge or ISFET in the NanoBridge or ISFET array. In further embodiments, the bias voltage difference may be compensated for by software or firmware, wherein the effect of the voltage bias may be measured, mapped and said map is used to compensate for variations in the signal level from the array of NanoBridges or ISFETs.

In some embodiments, a reference electrode is used to bias the reagent fluid relative to the sensor electrodes or active area of the sensor device, which may be an array of NanoBridges and/or NanoNeedles, or ChemFETs. In some embodiments, the reference electrode is configured to be a part of sensor device. In further embodiments, there are multiple reference electrodes, where one or more of the reference electrodes is part of or associated with a flow cell associated with the sensor device. In other embodiments, two or more reference electrodes are associated with the sensor device. In some embodiments, multiple reference electrodes maintain a substantially similar reagent voltage at all of the members of the array, which might be difficult in a flow cell where the fluidic thickness is sufficiently thin as to allow significant resistance over the surface of the sensor array.

In some embodiments, at least one additional electrode may be provided to bias the bulk reagent solution in the flow cell. This electrode could be the same electrode(s) used at other times to concentrate sample and/or other reagents. In some embodiments the voltage impressed on the electrode(s) may be used to bias the detectors at an optimal point in their response curve, for example, to provide appropriate offset to optimize the amount of gain, which may provide maximal signal within the available dynamic range of an analog to digital converter, so that A/D quantization error may be minimized.

In some embodiments, the bias level may be modified as a sequence reaction proceeds, and the amount of charge which is proximate to a sensor changes. In some embodiments a reading may be taken using the sensors, after which the bias level may be changed, followed by reading the sensors again, so that any nonlinearity or unexpected offsets which result from changing the bias voltage may be observed and

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 compensated for by the software. In some embodiments, positive charge is provided on or near the bead or colony, such that the sensor may be biased to an appropriate level.

In some embodiments, multiple reference electrodes may be used with NanoNeedle sensors, NanoBridge sensors, ISFET sensors, or ChemFET sensors.

Electronic sensors, such as ChemFETs may be designed to have a wide dynamic range, as is the case with some pH sensors. They may alternatively be designed to have a smaller dynamic range, but higher sensitivity. In one embodiment, both the dynamic range of the sensor and the sensitivity of the sensor is optimized, by including an additional element to the system which biases the active region. Said element may be a reference electrode or electrodes, wherein a variable voltage may be impressed between the reference electrode(s) and the active area of the sensor(s) (e.g., ChemFET or NanoBridgc). Adjustment of the voltage can permit highly sensitive detection despite a wide change in the amount of charge interacting with the sensor. For example, a sensor may be optimized to work with a sequencing reaction where the target DNA is 100 base pairs long. Alternatively, if the target DNA is 1000 base pairs long, the sensor may no longer be working within the sensor’s dynamic range. The voltage between the reference clcctrodc(s) and the active area may then be adjusted so that the sensor is permitted to work within its dynamic range. If in the course of the sequencing reaction, the extended primer has been extended to 500 base pairs, the sensor may again no longer be within its dynamic range. The reference voltage may again be modified to bring the sensor within its dynamic range. Additionally or alternatively, a back gate may be used in much the same fashion. In a further improvement, the back gate may be segmented, such that there may be different sections of the back gate for different areas of a sensor array. There may be many sections, so that it is possible to have an individual back gate for each sensor, permitting compensation for different sequence dependent rates at which the primer is extended.

In some embodiments, reference voltage(s) are changed when employing NanoNeedle sensors, NanoBridge sensors, ISFET sensors, or ChemFET sensors.

In some embodiments, measurements of polymerase incorporation are performed to determine the sequence of a DNA target. Multiple measurements may typically be needed in order to insure that the profile of incorporation is properly captured and measured, for example to determine the number of bases which have been incorporated in a homopolymer run. Such a measurement may measure byproducts of the incorporation reaction, such as PPi or hydronium ions. For a large array of sensors, such a measurement may require very high data collection rates, which may challenge the sensitivity of the sensor, preventing insufficient signal to noise to provide desired error rates associated with the sequencing data. There may be difficulties associated with trading off the errors associated with phase error, and thus length of read, and the

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 errors associated with accurately measuring which base, and how many bases have been incorporated. This may be a result of needing a low ionic concentration for sensor accuracy, and much higher concentrations in order for the polymerase to function accurately without phase errors. Thus, in some embodiments, two or more different reagent conditions are used during sequencing, where at least one set of reagent conditions is optimized for polymerase accuracy and minimization of dephasing, and a second reagent is optimized for detection, for example by having a very low ionic strength. Reading the sensor separately from the incorporation event may improve the sequencing data accuracy and read length. In some embodiments less data is required as the sensor may no longer be forced to be read at a high data rate to capture the polymerase incorporation event, but may instead be read a small number of times, potentially as few as a single time. The electronics may also have time constants which may be sufficiently long to permit sensor noise to be significantly reduced. Furthermore, in some embodiments, the reduced data requirements may simplify the data processing hardware, data transfer requirements, and data storage requirements.

In some embodiments, read buffer may be of lower ionic concentration than would be optimal for use for polymerase enzymes. In some embodiments, the ionic concentration of the read buffer may be one third the ionic concentration of the incorporation buffer; or in other embodiments, the ionic concentration of the read buffer may be one third to one tenth the ionic concentration of the incorporation buffer, one tenth to one thirtieth the concentration of the incorporation buffer, one thirtieth to one hundredth the ionic concentration of the incorporation buffer, or one hundredth to one thousandth the ionic concentration of the ionic concentration of the incorporation buffer.

In some embodiments, the pH of the incorporation buffer and the read buffer may be substantially the same pH. In other embodiments, the pH of the incorporation buffer and the read buffer may be noticeably different, for example, where the pH of the incorporation buffer is optimized for optimal activity and or accuracy of the polymerase enzyme, such as pH 8.5, while the read buffer is a pH that minimizes the conductivity of the read buffer, such as pH 7.0 (e.g., where the concentration of OH- and H+ are the same at IO'⁷ molar). In some embodiments, the optimal pH for minimal read buffer conductivity is slightly higher than pH 7.0, as the mobility of OH- is lower than that of H+. Thus in some embodiments, the pH of the read buffer is between pH 6.5 and pH 8.0, between pH 6.8 and pH 7.5, or between pH 7.0 and pH 7.2, while the pH of the incorporation buffer is between pH 7.5 and pH 9.0, between pH 8.0 and pH 8.8, or between pH 8.3 and pH 8.5.

In some embodiments, different reagent buffers are used with NanoNeedle sensors, NanoBridge sensors, ISFET sensors, or ChemFET sensors.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In some embodiments, an integrator is incorporated with the sensor to maximize the amount of time given to each sensor in order to reduce the read noise of each sensor. The integrator may include a capacitor associated with each sensor in the array. In other embodiments, the sensor is configured as a capacitive sensor, where there is no current flow, but rather an accumulation of charge during a chemistry cycle. In some embodiments, either an integrating device or a capacitive device, the sensor may have local amplification electronics for each pixel. In other embodiments, the charge is moved, in a manner similar to a CCD to a readout port.

In some embodiments, integrators are used as a part of the sensor, where the sensor comprises NanoNccdlc sensors, NanoBridge sensors, 1SFET sensors, or ChemFET sensors.

There may be one or more readout ports associated with each device. In some embodiments, each comer of the device may have a readout port; in other embodiments, there may be many ports along opposite sides of the device, allowing a reduced readout rate, and associated improved signal to noise. In other embodiments the readout circuitry can divide the array into columns or rows. In other embodiments the readout circuitry can be placed under channel support or channel separation features. In further embodiments, there may be multiple sets of readout circuitry, where the array of sensors is divided into multiple subarrays, and the multiple sets of readout circuitry are positioned such that the readout circuitry is coincident with the channel support or channel separation features. In some embodiments, it may be desirable to use flow cells with minimal reagent volume; as such it is desirable to have the height of the flow cell be as short as possible. For example, it may be desirable for the flow cell to be 300 microns tall or less, 100 microns tall or less, or 50 microns tall or less. In some embodiments it may be desirable to use a semiconductor device, which may be one centimeter square or larger, potentially as large as 10 centimeters square. A flow cell that is wide enough to cover a significant amount of the width of the sensor chip may have significant difficulties with mechanical tolerances due to flatness of one or both major surfaces of the flow cell with respect to the other surface, particularly if one surface is a molded plastic part or a PDMS or similar polymer part. As a result it may be desirable to use support posts, channels or other support shapes to prevent flatness tolerances from collapsing or expanding beyond desired tolerances.

In some embodiments, the system may use a sensor, such as a bridge sensor, which is arranged in a manner similar to a Fin FET, whereby two or three sides of the channel may be accessible to interact with the surroundings, such as, for example, DNA which is bound to the surface of the channel. The sensor channel may have a vertical dimension perpendicular to the substrate which is greater than the horizontal cross

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 section of the channel. Such a device may have greater sensitivity than a device which has but a single surface accessible to the sample.

In other embodiments, it may be desirable to use a material to provide more surface area than may be available with a planar or polished planar electrode. In some embodiments it may be desirable to use black platinum, platinum metal sponge, or a platinized metal, which may be platinized platinum, platinized titanium, platinized irridium, platinized Niobium, platinized tantalum, platinized zirconium, or other platinized metals as an electrode material. Said electrode may be a reference electrode or may be an electrode as part of a NanoNeedle. In other embodiments, the electrode surface is fabricated of other members of the platinum metal group: palladium, rhodium, ruthenium, iridium, or osmium, which may be used in the same manner as platinum to form an electrode surface with much higher surface area than a planar or polished electrode would form.

In some embodiments, the process of platinization may include cleaning a support material, potentially utilizing aqua regia, HC1, and HNO3, followed by a plating process which may utilize chloroplatinic acid and lead acetate.

In other embodiments, the electrode surfaces may include iridium oxide, titanium nitrate, or polypyrrole/poly(styrenesulphonate) conducting polymer. Fabrication of said iridium oxide may be effected by sputtering using standard photolithographic processes. Malleo et al (Review of Scientific Instruments 81, 016104) describe the increase in the effective interfacial capacitance of different materials relative to a bright platinum electrode as ranging from 240 times higher for Platium black, 75 times higher for iridium oxide, and 790 times higher for polypyrrole/'poly(styrenesulphonate) conducting polymer.

In some embodiments, sensors with larger surface areas arc used with NanoNeedle sensors, NanoBridge sensors, ISFET sensors, or ChemFET sensors.

In some embodiments, a NanoNeedle, NanoBridge, ChemFET or ISFET is fabricated such that the sensor is created on the surface of a substrate such as silicon, fused silica, glass or other similar material. In other embodiments, the sensor is fabricated such that it projects vertically or horizontally above the substrate, such that the sensor is more accessible to the fluid and reagents. The greater accessibility to fluid and reagents may decrease the time needed for a sequencing reaction to occur, allow lower concentrations of reagents to be used, and increase the sensitivity of the sensor by increasing the surface area associated with the active area of the sensor.

In some embodiments, electrode(s) may be fabricated using an angled rotated deposition approach, which may employ glancing angle deposition as described by Zhao et. al. (p59-73 SPÏE Vol 5219 Nanotubes and Nanowires), or may be fabricated using the PVD methodology described in US 6,046,097.

Figure 26 shows data from a one run of a NanoNccdlc sequencing reaction, wherein run data is scaled and shown relative to a linear plot of base incorporations. dNTPs which should not incorporate are shown as mostly overlapping previous data, and multiple incorporation base data is shown as having quite good linearity (R2 = 0.9974).

In alternative embodiments, the system or method detects kinetics of single molecule reactions, such as an enzymatic reaction. In some embodiments, the reaction may a hybridization reaction, whereby a bead or particle with a hybridization probe attached thereto may be caused to be held in place above a sensor, and the change in charge proximate a sensor or sensors resultant from a hybridization reaction may be measured. In an alternative embodiment, the hybridization probes may be attached on or proximate to the sensors, whereby the change in charge resultant from the progression of a hybridization reaction may be measured. In some embodiments, an electric field may be used to concentrate DNA from a reagent solution into the volume w here the hybridization probes are attached, which may be on a bead or particle, or may be on or proximate to said sensors. Said electric field may be a DC field, an AC field, or a combination thereof.

In some embodiments, a real time PCR reaction is monitored by using the sensor or sensors to monitor the change in conductivity or change in charge present, resulting from the incorporation of dNTPs into amplicons, and/or the release of pyrophosphate and hydronium ions with higher mobilities. In an alternative embodiment, an isothermal reaction amplifying target DNA is detected by the resultant change in conductivity from the incorporation of dNTPs into amplicons. In other embodiments, the sensors monitor the progression of an Immuno-PCR reaction, where a sandwich assay captures an antigen, and the detector antibody has a probe DNA oligo attached thereto, whereby a realtime PCR assay may then detect and quantify the presence and quantity of antigen, by the detection of the change in conductivity or charge as previously described. In another embodiment, an isothermal reaction detects and quantifies an antigen of interest.

In some embodiments, protein detection may be cffcctualizcd by direct measurement of the reaction, by measurement of a sandwich assay, or by measurement using an aptamer, or by other appropriate methods wherein a change in counter ions or a change in charge associated with the target which may be bound, attached or associated with the sensor.

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

In another embodiment, a nucleic acid aptamer which is bound to or proximate to the sensor or sensors is used to detect the presence and quantity of a target. The aptamer may bind to the target, changing the charge which may be detected by said sensor as previously described. In alternative embodiment, the aptamer is attached on or proximate to said sensor or sensors, and an increase in the conductivity is detected as a result of binding of a target thereto.

In a further embodiment, blunt end ligation may be performed with ligands that have different binding reagents on the 3’ and 5’ ends of said ligands. The electrodes of the NanoNeedle may be conjugated with the complementary reagents for binding e.g. the 3’ end of the ligands may have a thiol group, and one electrode may be fabricated of gold, while the 5’ end of the ligands may have a PNA sequence, and second electrode may have the complement to said PNA sequence. The strand of DNA may then be electrophoretically and or dielectrophoretically concentrated to the area of the NanoNeedle, where said DNA strand may then bind with one end associated with one electrode, and the other end associated with the second electrode of the NanoNccdlc. Polymerase and primer may be bound to the DNA strand, or may be introduced later. Measurement of incorporation events may then result from direct measurement of the impedance of the DNA combined with the much larger conductivity of the counter ions associated with the DNA.

In some embodiments, the sensor device, which may be a NanoBridge or a NanoNeedle, generates digital output data. Said digital output may comprise any of a number of output physical/Data link/protocol formats, including USB2, USB3, Firewire, Gige, single link or dual link DVI, HDMI, S/PIF, ADAT lightpipe, AES3, MADI-X, I²S, AC’97, MC’97, McASP, S-Video, ATM, SONET, SDH, UTP, STP, AUI, HDLC, 802.1, ARP, VLAN, HDLC, ATM, Frame Relay, Q in Q, PPP, BSC, DDCMP, Banyan, CDMA2000, DECnet, CDPD, FUNI, CDMA, X.25, GPRS, GR-303, H.323, NFS, ISDNSS7, TCIP, UMTS, WAP,XNS, MDLP, Infiniband, amongst many others.

Said output may be in compressed format, such as an MPEG 1, MPEG 2, MPEG4, DVA, AVI, MOV, MPG, Video CD, RM, WMA, WMV, WAV, FLC, FLI, BMP, PCX, TGA, TIF, JPG, PCT, GIF, Flash, QuickTime, MP3, or sequences thereof.

Said sensor device may be configured to have more than one digital I/O connection, and may have more than one output format; for example, one digital connection may be used to control the operation of said sensor, while one or more digital connection sends data from the sensor to additional device which may be part of an instrument of which the sensor is a part. Said additional device may be a data storage, device, or may be a computational device. Said additional device may be a GPU, or a set of GPUs such as a GPU array.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Said data may be transferred directly from said sensor to a hard disk, directly from said sensor to a solid state drive, or directly from said sensor to a GPU, or directly from said sensor to a GPU cluster, GPU blade, or GPU server, a CPU, or the memory associated with a GPU, or CPU. In some embodiments, an instrument or system may have more than one sensor. In such an instrument or system, data may be accumulated from more than one sensor, and thence sent directly to a hard disk, a solid state drive, a GPU, a GPU blade, a GPU server, a GPU cluster, a CPU or the memory associated with a GPU or CPU.

In some embodiments, a single sensor may have more than one digital output. In other embodiments the digital output may be configured to directly connect to another part of the system, such as a solid state drive or a memory associated with a GPU or CPU, wherein two or more parts of the system may be a part of a MCM (Multi Chip Module), or SIP (System in package). Said MCM may be a laminated MCM, a deposited MCM, a ceramic substrate MCM or a chip stack MCM. The sensor may be a part of the MCM, or may be separate from said MCM. Said sensor may be configured such that said sensor may be removed and a second sensor may be utilized. Said sensor may be interconnected using a socket; said socket may be a zero insertion force socket for a PGA (Pin Grid Array), a LGA (Lan Grid Array) socket, or a slotket.

The data may be compared with data in a CAM (Content Addressable Memory), or a CAM memory which permits a selectable number of errors in DNA mapping, such as a ternary CAM. Said CAM memory may have multiple levels in a manner similar to that of TLB (Translation Lookaside Buffers), wherein one level of said CAM or TLB may be faster and smaller than another level of said CAM or TLB.

To improve the sensitivity of either the NanoNeedle or the Nanobridge, a local amplifier may be provided. The amplifier may be either a BJT or an FET. The sensor can be fabricated as a narrow structure, and can be etched under the structure so that both sides are accessible to changes in pH, conductivity or local charge. The surface of the device may be rough, permitting greater surface area for binding of sample molecules. Electric confinement of ions may be effected as will be described further hereafter.

In some embodiments of the current invention, the image sensor array may use amplifier designs similar to those in CMOS active pixel image arrays; these may include three transistor, four transistor, five transistor, or six transistor circuits, depending on the signal to noise needed, and whether a global shutter equivalent is desired if a integrating circuit is utilized. Said amplifier structure may be arranged in a one to one correspondence with said image sensor array, potentially providing significantly better signal to noise than might otherwise be possible utilizing a common amplifier for multiple sensors.

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

Integrated Systems

The invention further provides methods and systems for localizing samples and reagents into a volume where a desired reaction or binding may occur. The invention is this aspect may eliminate or reduce the need for whole genome amplification, and thus reduce the coverage needed.

In some embodiments, the DNA sequencer may be part of a larger system, where more portions of the workflow are automated. These portions of the workflow which may be automated may include cell lysis, DNA purification, DNA amplification, DNA library preparation, colony generation, sequencing, primary analysis and base calling, mapping of sequences to a reference, and determination of whether a genetic disease or other genetic characteristic is present. In some embodiments, the system may have more functionality, including a means to sort cells, such as cancer cells from blood, utilizing a flow' cytometer or affinity pullout of desired or undesired cells.

It may be desirable to process multiple samples in a single chip, since many projects do not require the full capacity of a chip. Other projects may have a single sample that would exceed the capacity of the chip. In some embodiments one or more samples could be introduced into the instrument in individual tubes, tube strips, 96-well plates, 384-well plates, etc. In some embodiments the sample wells could be sealed to prolong life on the instrument. In other embodiments the plates may be cooled to increase sample life. In other embodiments the samples could be accessed in a software selectable manner by a robotic pipettor.

Prior to amplification the beads will need to be loaded with a single DNA fragment in order to create monoclonal beads. Typically the DNA concentration is determined and then it is introduced to beads in a dilute form so that on average less than 1 fragment will bind to each bead. Many beads have zero DNA fragments, fewer have a single fragment and a small number have 2 or more fragments. The steps needed for quantitation often require a separate instrument and separate processing.

In one embodiment a target concentration is created by a hybridization based pullout. A solid support such as pull-out beads may be functionalized with a controlled number of binding sites. In some embodiments these are DNA primer complements. The unamplified sample may have known primers ligated on each end. In some embodiments the primers may hybridize to the DNA on the pull-out beads. After the hybridization sites are fully occupied residual DNA may be washed away, and the DNA bound to the beads may subsequently be denatured releasing a known quantity of DNA.

In another embodiment the primers ligated to each DNA fragment are bound to the primer complement and detected using fluorescence detection of an intercalating dye. Since the primers are of a known length, the signal level will be proportional to

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 the number of fragments. In another embodiment polymerase and associated dNTPs could be introduced creating full length double stranded DNA. When combined with the information from the primer signal the full length intercalating dye signal level would then allow determination of the mean fragment length.

In another embodiment dielectrophoresis is used to concentrate DNA. During or after concentration the electrical current is measured to determine the DNA concentration. In another embodiment the concentrated DNA is quantified by the use of intercalating dyes as described above. In another embodiment, the concentration of the DNA is determined directly by optical absorbance. The optical absorbance determination may, for example, use an optical source which produces light at 260nm.

In one embodiment the sample is made very dilute and/or a small volume of sample reagent and loaded onto beads. DNA would bind to some of the beads and then be amplified in the virtual reactors creating beads with DNA. The sequencing primer may be made shorter than the complement ligated to the sample DNA. Since the sequence is known, the correct dNTPs could be added and detected. In one embodiment multiple dNPTs are simultaneously added. For example, if all dNTPs are added the polymerase would extend to the end of the fragment generating a large signal. Said large signal could be generated as a part of the amplification process. This may allow the detection and counting of the number of beads that have DNA even if the beads had minimal amplification. Knowing how many beads have signal may allow calculation of the proper dilution to generate the ideal number of monoclonal beads.

Similarly, measurements made using electrical current, optical signals, or other signals which indicate the concentration of DNA in the sample may be used to determine the dilution level, if any, needed to optimally utilize the DNA in the system.

In some embodiments, dilution is needed to properly generate colonies. Similarly dilution may be needed for a nanopore system in order to prevent pore clogging, and conversely, to optimize the duty cycle whereby a pore may be occupied with a DNA strand. Dilution may be effectualized as part of an emulsion PCR system, a bridge PCR system, a nanopore sequencing system, or a single molecule optical system.

In other embodiments, concentration may be implemented as part of system, and may be effectuated by dielectrophoresis, hybridization, ethanol precipitation or other methods, and may be used to increase the concentration of DNA to improve an emulsion PCR system, a bridge PCR system, a nanopore sequencing system, or a single molecule optical system.

A primary system may determine the concentration of template DNA using software or hardware to make said determination, and may then either concentrate or

100

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 dilute as needed prior to utilizing said template DNA in the next appropriate step said system, which may be amplification or sequencing. Said determination step may also make use of a prior calibration step, which may use standard comprising DNA of a known concentration, or may use DNA of an initially unknown concentration, where the concentration is determined by a separate system. The determined concentration may be entered transferred, or otherwise communicated to said primary system. Said primary system may store any values needed for calibration locally in the primary system, or may store it in a part of a larger system, or in a separate computer, or in a data base. Said calibration information may also include additional information, such as the time of calibration, the operator, the sample or standard utilized for calibration, or other information as may be determined to be relevant.

Many current systems use whole genome amplification in order to have sufficient DNA for their protocol. Typical amplification methods may use degenerate primers and PCR, random hexamers and isothermal amplification or other methods for amplification of genomic DNA. Said amplification may amplify genomic DNA by a thousand fold or more. This amplification can introduce bias and is an additional cost in time and resources. The ability to reduce or eliminate the need to amplify the sample is desirable. In one embodiment the beads to be loaded are enclosed in a packed bed and sample is pumped across it. The sample can be pumped through the bead bed multiple times to provide additional opportunities for the sample to bind. The high surface area to volume should allow for minimal sample to be used. The beads can subsequently be moved into a flow cell whereby they may be held in place by a magnetic array, and local colonies may be created on the beads by PCR or isothermal amplification.

In another embodiment the sample is concentrated in the amplification region using the existing electrodes of the emulsion free nano-reactor. In one embodiment electrodes may be established on a single plane. In another embodiment electrodes may be added to a second plane parallel to the plane of the virtual reactors. In other embodiments mixtures of AC and DC voltage inputs are anticipated.

In other embodiments, whole genome amplification or targeted amplification, such as amplification which targets the exome, the conserved regions of the genome, a cancer panel, or other targets of interest may be implemented as part of a subsystem within an integrated system. Said targeted amplification may also incorporate barcodes for different samples as a part of the amplification process. The amplified DNA may then have its concentration determined as explained herein, prior to undergoing clonal amplification for subsequent sequencing, using a clonal sequencing subsystem and method, which may be a part of the integrated system. Alternatively or addition, the

101

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880

DNA is sequenced directly using a single molecule sequencing subsystem and method which may be a part of said integrated system.

Since many projects may not require the full use of a sequencing chip or flow cell it may be desirable to load multiple samples into different portions or areas of a single chip or flow cell. In one embodiment, samples are directed into separate zones separated by walls on the chip or flow cell using valves integrated into the chip or flow cell assembly. Such valves may be PDMS valves integrated into the fluidic path. In another embodiment there may be separate zones with separate inputs and outputs. In another embodiment samples may be directed into separate zones on a chip or flow cell using a local electric field. A positive field may be applied to attract DNA to desired regions, while a negative field may be applied to repel DNA from undesired regions. In another embodiment samples may be directed into separate zones using electromagnets to control the positioning of magnetic or paramagnetic beads. In another embodiment samples may be delivered into individual lanes using self sealing ports. Self sealing ports can include rubber septa and needles.

In another embodiment samples can be injected at different time points and new beads and bead locations can be distinguished using sensor signals relative to that previously determined for said sensors, where the bead locations where previously empty.

In a further embodiment, electrowetting or optoelectrowetting is used to deliver samples to distinct and separate regions of a chip or flow cell.

In some embodiments, containers for the reagents may be cooled as needed, for example, regents which contain samples, polymerase, phosphatase, or other enzymes may need cooling, for instance, to about four Celsius.

In some embodiments, the amount of reagent contained in the lines leading from the reagent containers to the valve manifold may contain a volume which is significant relative to that which is needed to perform a sequencing reaction. In order to prevent needing to discard this reagent, for example, at the beginning of a new sequencing run, it may be desirable to cool the lines from where they interface with the reagent containers, up to, or close to, where they enter the valving manifold. In some embodiments, where the valving manifold is sufficiently separated from where reagents enter the flow cell where sequencing or another reaction occurs to permit the flow cell and the valving manifold to operate at different temperatures, for example, about four Celsius and 20 to 40 Celsius respectively, it may be desirable to cool the valving manifold as well.

In some embodiments it may be desirable to use a system capable of more than one method of sequencing, wherein one method, process or subsystem for sequencing

102

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 provides information of one type, and a second method, process or subsystem sequencing may provide information of a second type. For example, in some embodiments it may be desirable to provide one method where the type of information may elucidate the structure of the DNA sample, enabling sequence reads which may span the length of repeat sequences, such as simple sequence repeats, short tandem repeats, microsatellites, minisatellites, variable number tandem repeats, interspersed repeats such as LINE repeats, SINE repeats such as Alu repeats, direct repeats, or inverted repeats, or other types of repeat sequences which may prevent proper complete assembly of a genome or other desired sequence or sequences of DNA. In some embodiments, it may be desirable to use a method, process or subsystem for sequencing which may, for example, elucidate the structure of DNA, where it may not be necessary to determine the sequence w'ith high accuracy. Tn some embodiments, it may be desirable to have association of reads which may be separated many bases apart, as is done in some systems, for example by mate pair sequencing or strobe sequencing. In some embodiments it is desirable to use a method, process or subsystem for sequencing which provides sequencing reads which have very high accuracy to detect, for example, single nucleotide polymorphisms, but for which a long sequencing length of read is unneeded. In some embodiments, it may further be desirable to use a method, process or subsystem for sequencing provides the ability to provide many short reads with low accuracy, as may be needed, for example, for whole transcriptome analysis.

In some embodiments, it is desirable to use different methods for a single sample, enabling, for example, detection of both single nucleotide polymorphisms and structural rearrangements from a single sample. In some embodiments, it is desirable within a single system to separate purified nucleic acids into two or more aliquots, which may then have different corresponding library preparation methods, subsystems, or processes which may include amplification, and may utilize different amplification methods, subsystems and processes as appropriate for the different desired sequencing methods, subsystems, or processes. The size, concentration, or volume of the different aliquots may be similar, the same or different, and may be different as appropriate for the different sequencing and/or library preparation method(s), subsystem(s), or process(es) which may be utilized to effect the different desired sequencing results. In some embodiments, the method may be different for different aliquots, but the subsystems used may be the same, and or the subsystem used for the different methods may be the same subsystem, wherein first one method is used, and then a second or more subsequent method(s) may be performed used the same subsystem. For example, it may be desirable to use different fragment lengths for different sequencing methods, wherein, for example, long fragments may be desirable for determining sequence structure, whereas short(er) fragments may be desirable for determination of single nucleotide polymorphisms. Thus it may be desirable to fragment different aliquots of

103 the sample to different average fragment lengths, wherein the average fragment length of one aliquot may be longer, potentially significantly longer than another aliquot. In some embodiments, the method of fragmentation may be the same, and may use for example, a sonicator, but the time the sonicator applies energy to said aliquots and/or the power level of the sonicator applied to said aliquots may be different, such that the level of fragmentation of the sample in the two or more aliquots may be different, potentially substantially different, and the resultant fragment length may be different, potentially significantly different. In other embodiments, it may be desirable to use a single system to generate long fragments and short fragments; in a further embodiment, it may be desirable to fragment DNA to size appropriate for long fragments, remove an aliquot, and further fragment remaining DNA to a size appropriate for said short fragments.

In other embodiments, one aliquot of nucleic sample may be genomic DNA for which single nucleotide polymorphisms may be determined, and a second aliquot may be RNA for which a transcriptome may be desired. Amplification methods, subsystems, or processes may be different for the genomic DNA and the RNA, wherein a conversion from RNA to cDNA may be effected, and wherein the amplification protocol may be different for the different aliquots, as the accuracy needed for amplification for single nucleotide polymorphisms may be much higher than the accuracy needed for conversion of RNA to cDNA and subsequent amplification of said cDNA. In some embodiments it may be desirable to use lower concentrations and/or less expensive reagents for amplification of cDNA for transcriptome analysis than for amplification of genomic DNA for single nucleotide polymorphism analysis. In further embodiments, it may be desirable to use shorter cycle times for amplification of cDNA for transcriptome analysis than for amplification of genomic DNA for single nucleotide polymorphism analysis, which may speed time to answer allowing for transcriptome analysis to commence using the same sequencing subsystem which may be used subsequently by said genomic DNA single nucleotide polymorphism analysis. Any other combinations of separation of nucleic material for subsequent analysis using different sequencing methods, subsystems or processes may be envisioned.

In some embodiments, different types of sequencing detection methods, subsystems or processes are used. For example, one subsystem may use single molecule sequencing as described by Church et al in US5,795,782, Hancck ct al in US US8,137,569 Korlach ct al in US7,361,466, and Clark ct al in US2011/0177498, which may have low accuracy and may have very long sequence reads, while another subsystem might use optical or electrochemical detection of sequencing by synthesis as described by McKernan et al in US2009/0181385, Balasubramanian in US6,833,246, Nyren et al in US6,210891,Bridgeham et al in US7,282,370, Williams ct al in US7,645,596, Rothberg 104

CA 2837486 2018-10-02 et al in US7,948,015, Toumazou et al in US8,114,591, and Miyahara et al in US7,888,013.. Thus in some embodiments, a single system might have at least two different detection subsystems, wherein said two different detection subsystems may use different sequencing methods, sequencing detection methods, or sequencing processes, and wherein said different sequencing methods, sequencing detection methods, or sequencing processes may be performed at the same time either for the same sample or for different samples, or may be performed at different times for the same sample or different samples.

Exemplary integrated systems are illustrated in accompanying drawings.

Figure 1A depicts a complete sequencing system 100, which may comprise an external computing device 102, and an integrated system 104. The integrated system may comprise a rack module 110, which may further comprise a fluidics interface subsystem 116, a set of sequencing/sample prep cards 112, and individual sequencing subsystems 114 on each sequencing/sample prep card 112. Schematic of sequencing/sample prep 120 includes library prep 122, re-useable magnetic arrays 124, which may further comprise sequence detectors 126, which result in sequencing data 128.

Figure IB depicts a complete library prep subsystem 130, which includes sample cell input 131, cellular lysis and protein removal 132 resulting in un-fragmented genomic DNA 133, which may be input to a fragmentation and separation subsystem 134, which may then output fragmented genomic DNA 136, which may be transported along with a set of beads 135, to a virtual well array 137 for amplification, and then said beads may be separated utilizing a field 138 in an bead enrichment module 139 into sets of beads with amplification from set of beads without amplification.

Figure IC schematically illustrates a genomic DNA fragmentation and separation system 140, comprising input un-fragmented genomic DNA and fragmentation beads 142, which are input to a fragmentation subsystem 144 wherein said un-fragmented genomic DNA may be fragmented. The fragmented DNA may be separated by size in a channel 146 using pumping or electrophoretic force from pumps or electrodes 147, and may then be moved to an output from said separation channel 146 via fluidic outputs 148, and outputting said fragmented DNA 149.

Figure ID shows an embodiment of a PDMS library preparation module 150 which includes a lysis section 152, a protein removal section 154, and an amplification section 156.

While various embodiments have been described above, it should be understood that they have been presented by way of example only, and not limitation. Where

105

CA 2837486 2018-10-02

CA 02837486 2013-11-26

WO 2012/166742

PCT/US2012/039880 methods and/or schematics described indicate certain events, and/or flow patterns, and/or chemical reactions occur in a certain order, the ordering of certain events and/or flow' patterns and/or chemical reactions may be modified. While the embodiments have been particularly shown and described, it will be understood that various changes in 5 form and or detail may be made.

Although various embodiments have been described as having particular features and/or combinations of components, other embodiments may be possible having a combination of any features and/or components as discussed above.