Methods for sequencing polynucleotides

Abstract

The present invention relates to an improvement in methods for high-throughput nucleic acid sequencing, in particular to an improvement in methods for performing extension reactions during pairwise sequencing. The present invention relates to a method for performing a strand resynthesis extension reaction during pairwise sequencing, wherein the strand resynthesis extension reaction is performed between a first sequencing read and a second sequencing read, the strand resynthesis extension reaction extends one or more immobilized primers to copy a first template strand to generate a second immobilized template strand, and the strand resynthesis extension reaction is performed using a non-thermostable strand-displacing polymerase at a temperature below 55°C, preferably 38°C.

Description

[Technical Field]

[0001] The present invention relates to improvements in methods of high-throughput nucleic acid sequencing, and in particular to improvements in methods for performing extension reactions during pairwise sequencing. [Background technology]

[0002] Nucleic acid sequencing methods have been well known in the art for many years. Some such methods are based on the successive incorporation cycles of fluorescently labeled nucleic acid analogs. In such "sequencing by synthesis" or "cycle sequencing" methods, the identity of the added base is determined after the addition of each nucleotide by detecting the fluorescent label.

[0003] In particular, U.S. Patent No. 5,302,509 describes a method for sequencing a polynucleotide template, which involves performing multiple extension reactions using a DNA polymerase or DNA ligase to successively incorporate labeled polynucleotides complementary to the template strand. In such a "sequencing by synthesis" reaction, a new polynucleotide strand paired with the template strand is constructed in the 5' to 3' direction by successively incorporating individual nucleotides complementary to the template strand. The substrate nucleoside triphosphates used in the sequencing reaction are labeled at the 3' position with different 3' labels, allowing for the determination of the identity of the nucleotides incorporated as consecutive nucleotides.

[0004] To perform accurate sequencing, reversible chain-terminating structural modifications or "blocking groups" can be added to substrate nucleotides to ensure that nucleotides are incorporated one at a time in a controlled manner. As each single nucleotide is incorporated, the blocking group prevents the incorporation of any further nucleotides into the polynucleotide chain. Once the identity of the last incorporated labeled nucleotide is determined, the label moiety and blocking group are removed, allowing the next blocked, labeled nucleotide to be incorporated in the subsequent round of sequencing.

[0005] The techniques of "paired-end" or "pairwise" sequencing (such terms are used interchangeably herein) are generally well known in the art of molecular biology, particularly in the context of whole-genome shotgun sequencing. Paired-end sequencing allows for the determination of two "reads" of sequence from two locations on a single polynucleotide duplex. The advantage of the paired-end approach is that significantly more information can be obtained from extending each of "n" bases twice from a single template than from sequencing "n" bases from each of two independent templates in a random manner. By using appropriate software tools for assembling sequence information, it is possible to use the knowledge that "paired-end" sequences are not completely random, but are known to occur in a single duplex and are therefore linked or paired in the genome. This information has been shown to significantly aid in assembling the entire genome sequence into a consensus base sequence.

[0006] Paired-end sequencing is typically performed using specialized circular shotgun cloning vectors. After cleaving the vector at a specific single site, the template DNA to be sequenced (typically genomic DNA) is inserted into the vector and the ends are resealed to form a new construct. The vector sequences flanking the insert DNA contain binding sites for sequencing primers, which allow sequencing of the insert DNA on the opposite strand. However, the need for sequencing primers at both ends of the template fragment makes array-based sequencing techniques very difficult to use. Array-based techniques, which typically rely on single-stranded templates, generally allow sequencing from only one end of the nucleotide template because the complementary strand is not attached to a surface.

[0007] In order to maximize the throughput of nucleic acid sequencing reactions, it is advantageous to be able to sequence multiple template molecules in parallel.The parallel processing of multiple templates can be achieved by using nucleic acid sequencing technology.These sequences are typically composed of a high-density matrix of polynucleotides immobilized on a solid support material, and generally rely on single-stranded templates.

[0008] Various methods have been reported for double-end sequencing of polynucleotide templates that can be carried out on solid support.For example, nucleic acid amplification methods are well known that allow amplification products to be immobilized on solid support to form arrays consisting of clusters or "colonies" formed by a plurality of identical immobilized polynucleotide strands and a plurality of identical immobilized complementary strands.The nucleic acid molecules present in the DNA colonies on the clustered arrays prepared according to these methods can provide templates for sequencing reactions.

[0009] WO 2008 / 041002 describes a pairwise sequencing method, which includes: (a) providing a solid support having a plurality of double-stranded template polynucleotides immobilized thereon, each formed from complementary first and second template strands linked to the solid support at their 5' ends and a plurality of copies of one or more 5'-end-immobilized primers capable of hybridizing to the 3' ends of the first template strands; (b) treating the plurality of double-stranded template polynucleotides such that the first template strands hybridize to the 5'-end-immobilized primers; and (c) performing a first sequencing read to identify the sequence of the template polynucleotides. (d) performing an extension reaction to extend one or more of the immobilized primers to copy the first template strand to generate second immobilized template strands; (e) processing the plurality of first immobilized template strands and the second immobilized template strands to remove the first template strands from the solid support; and (f) performing a second sequencing read to determine the sequence of a second region of the template polynucleotide, wherein determining the sequences of the first and second regions of the target polynucleotide achieves pairwise sequencing of the first and second regions of the target double-stranded polynucleotide.

[0010] There remains a continuing need to improve sequencing platforms, including increasing operational speed, improving efficiency, increasing accuracy, and / or simplifying the process. Summary of the Invention

[0011] According to one aspect of the present invention, there is provided a method for performing a strand resynthesis extension reaction during pairwise sequencing, wherein the strand resynthesis extension reaction is performed between a first sequencing read and a second sequencing read, the strand resynthesis extension reaction extending one or more immobilized primers to copy a first template strand to generate a second immobilized template strand, and the strand resynthesis extension reaction is performed using a non-thermostable strand displacement polymerase at a temperature below 55°C.

[0012] According to one aspect of the present invention, there is provided a method for performing an extension reaction during pairwise sequencing of a first region and a second region of a target polynucleotide, wherein the first region and the second region are in the same target polynucleotide, and the pairwise sequencing comprises the steps of: (a) providing a solid support having immobilized thereon a plurality of first and second double-stranded template polynucleotides, each formed from a first template strand and its complement or a second template strand and its complement, wherein the first and second template strands are linked to the solid support at their 5' ends, and the first template strand or the second template strand linked to the solid support further comprises a 5' immobilized extension primer; (b) selectively removing the first template strand complement and the second template strand complement, and selectively removing the second template strand to allow hybridization of a first sequencing primer to the first template strand; (c) performing a first sequencing read to determine the sequence of a first region of the template polynucleotide by sequencing-by-synthesis or sequencing-by-ligation techniques; (d) performing an extension reaction to extend one or more of the immobilized primers to copy the first template strand to generate a second immobilized template strand; (e) selectively removing the first template strand to allow hybridization of a second sequencing primer to the second template strand generated in step (d); (f) performing a second sequencing read to determine the sequence of a second region of the template polynucleotide by a sequencing-by-synthesis technique or by a sequencing-by-ligation technique, wherein determining the sequences of the first and second regions of the target polynucleotide achieves pairwise sequencing of the first and second regions of the target polynucleotide; The extension reaction of step (d) is carried out using a non-thermostable strand-displacing polymerase at a temperature below 55°C.

[0013] Surprisingly, it has been found that the extension reaction performed after the first sequence read can be performed at lower temperatures using a non-thermostable strand-displacing polymerase while still retaining a high level of strand resynthesis. The use of a non-thermostable strand-displacing polymerase may also provide additional advantages, including faster regeneration times, simplified protocols including fewer process steps, a reduction in the number of different reagents used and / or the amounts of reagents used, reduced chemical complexity, and / or a reduced complexity of the cartridge used to perform the method.

[0014] According to a further aspect, there is provided a method for pairwise sequencing of a first region and a second region of a target polynucleotide, wherein the first region and the second region are in the same target polynucleotide, and the pairwise sequencing comprises an extension reaction as described herein.

[0015] According to a further aspect, there is provided a method of improving data quality of a sequencing reaction comprising performing an extension reaction as described herein. [Brief explanation of the drawings]

[0016] [Figure 1] 1 shows an exemplary extension reaction workflow according to the present invention. [Figure 2] 1 shows an exemplary extension reaction workflow according to the present invention. [Figure 3] 1 shows an exemplary extension reaction workflow according to the present invention. [Figure 4] 1 shows an exemplary extension reaction workflow according to the present invention. [Figure 5] 1 shows the regeneration efficiency of the protocol according to the present invention. [Figure 6A] 1 shows an optimization analysis of a protocol according to the present invention. [Figure 6B] 1 shows an optimization analysis of a protocol according to the present invention. [Figure 7] FIG. 1 shows a schematic diagram of bridge amplification for paired-end resynthesis. DETAILED DESCRIPTION OF THE INVENTION

[0017] The following features apply to all aspects of the invention.

[0018] Sequencing generally involves four core steps: 1) library preparation to make multiple template polynucleotides available for sequencing, 2) cluster generation to make sequences of amplified single template polynucleotides on a solid support, 3) sequencing of the cluster sequences, and 4) data analysis to determine the target sequence.

[0019] The present invention relates to a regeneration step (also referred to herein as a strand resynthesis extension reaction or cluster regeneration) carried out between a first and second sequencing step during a pairwise sequencing process. The present invention relates to the use of a non-thermostable strand displacement polymerase in said extension reaction.

[0020] Typical steps of pairwise sequencing are well known and are described in International Publication No. 2008 / 041002, the contents of which are incorporated herein by reference. However, the key steps will be briefly described. Methods applicable to the present invention are described in International Publication No. 08 / 041002, International Publication No. 07 / 052006, International Publication No. 98 / 44151, International Publication No. 00 / 18957, International Publication No. 02 / 06456, International Publication No. 07 / 107710, International Publication No. 05 / 068656, U.S. Patent Application No. 13 / 661,524 and U.S. Patent Application No. 2012 / 0316086, the contents of which are incorporated herein by reference. Further information can be found in U.S. Patent Application No. 20060024681, U.S. Patent Application No. 200602926(U), WO 06110855, WO 06135342, WO 03074734, WO 07010252, WO 07091077, WO 00179553, and WO 98 / 44152, the contents of which are incorporated herein by reference.

[0021] Suitable templates for sequencing can be prepared by solid-phase nucleic acid amplification to generate nucleic acid colonies. The amplified (and subsequently sequenced) templates generally contain unknown regions flanked by known ends containing universal primers and are prepared, for example, according to the methods described in International Publication No. 07052006, the contents of which are incorporated herein by reference in their entirety. For example, templates can be derived from genomic DNA samples or cDNA libraries. Amplification can be performed using procedures similar to those described in International Publication Nos. 98 / 44151, 00 / 18957, 0206456, or 07107710, the contents of which are incorporated herein by reference in their entirety. When template nucleic acids are formed by solid-phase nucleic acid amplification, these non-target sequences can be derived from the primers used in the amplification reaction. Alternatively, non-target sequences can be attached to fragmented target sequences to incorporate them into nucleic acid molecules.

[0022] For example, suitable nucleic acids to be amplified with universal primers can be prepared by modifying a polynucleotide containing the target region to be amplified (and sequenced) by adding known adapter sequences to the 5' and 3' ends of the target polynucleotide to be amplified.

[0023] Adapters are typically short oligonucleotides that can be synthesized by conventional means. Adapters can be attached to the 5' and 3' ends of target nucleic acid fragments by various means (e.g., subcloning, ligation, etc.). More specifically, two different adapter sequences are attached to a target nucleic acid molecule, and the target nucleic acid molecule is amplified such that one adapter is attached to one end (i.e., the 5' or 3' end) of the target nucleic acid molecule, and another adapter is attached to the other end (i.e., the 3' or 5' end, respectively) of the target nucleic acid molecule. The resulting construct, which includes target nucleic acid sequences flanked by adapters, may be referred to herein as a "substrate nucleic acid construct" or a "template polynucleotide" or a "template construct." The template polynucleotide may be advantageously sized prior to modification with the adapter sequences.

[0024] Adapter sequences contain sequences that allow for the amplification of these molecules on a solid support to form a cluster using forward and reverse primers (e.g., S1 and S2) immobilized on the solid support (their general structure is described in more detail below). These sequences within the adapter are sometimes referred to herein as "primer binding sequences." To serve as a template for nucleic acid amplification, a single strand of the template construct must contain a sequence complementary to the forward amplification primer (e.g., S1), allowing the forward primer molecule to bind and prime synthesis of a complementary strand by copying the sequence complementary to the template and the reverse amplification primer (e.g., S2), which allows the reverse primer molecule to bind and prime synthesis of a second complementary strand. Sequences within the adapter that allow hybridization to immobilized primer molecules are typically about 20-40 nucleotides in length, although the invention is not limited to sequences of this length.

[0025] The exact identity of sequences S1 and S2 in the amplification primer, and therefore the cognate sequences in the adapter, is generally not critical to the present invention, so long as the primer molecule is able to interact with the adapter sequence to direct PCR amplification. The criteria for designing PCR primers are generally well known to those skilled in the art.

[0026] The template polynucleotide may also be prepared to further contain a tag or index sequence. For example, the use of tag sequences, such as those described in International Publication No. 05068656, the entire contents of which are incorporated herein by reference, allows for the analysis of multiple different samples in the same sequencing run while preserving the identity of each sample. The tag can be read at the end of the first read and / or the beginning (or end) of the second read. The present invention is not limited to two reads per cluster; three or more reads per cluster can be obtained by simply dehydrogenating the first extension sequence primer and rehybridizing the second primer before or after the strand resynthesis extension reaction. Methods for preparing samples suitable for indexing are described, for example, in U.S. Patent Application No. 60 / 899221.

[0027] In one example, the template polynucleotide comprises, from 5' to 3', a first primer binding sequence (e.g., P5), an index sequence (e.g., i5), a first sequence binding site (e.g., SBS3), an insert, a second sequence binding site (e.g., SBS12'), a second index sequence (e.g., i7'), and a second primer binding sequence (e.g., P7'). In another embodiment, the template comprises, from 3' to 5', a first primer binding site (e.g., P5', complementary to P5), an index sequence (e.g., i5', complementary to I5), a first sequence binding site (e.g., SBS3', complementary to SBS3), an insert, a second sequence binding site (e.g., SBS12, complementary to SBS12), a second index sequence (e.g., i7, complementary to I7), and a second primer binding sequence (e.g., P7, complementary to P7'). Either template is referred to herein as a "template strand" or a "template polynucleotide." The combination of the primer binding sequence, index sequence, and sequencing binding site may be referred to herein as an adapter sequence, and the single insert is flanked by a 5' adapter sequence and a 3' adapter sequence.

[0028] The sequence of the P5 primer binding sequence may comprise SEQ ID NO: 1 or a variant thereof, the sequence of the P5' adaptor may comprise SEQ ID NO: 3 or a variant thereof, the sequence of the P7 adaptor may comprise SEQ ID NO: 2 or a variant thereof, and the sequence of the P7' adaptor may comprise SEQ ID NO: 4 or a variant thereof. In embodiments, the variant has at least 80% sequence identity to SEQ ID NO: 1, 2, 3, or 4. More preferably, the variant has at least 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or at least 99% overall sequence identity to SEQ ID NO: 1, 2, 3, or 4. SEQ ID NO: 1: P5 sequence AATGATACGGCGACCACCGAGATCTACAC SEQ ID NO: 2: P7 sequence CAAGCAGAAGACGGCATACGAGAT SEQ ID NO: 3: P5' sequence (complementary to P5) GTGTAGATCTCGGTGGTCGCCGTATCATT SEQ ID NO: 4: P7' sequence (complementary to P7) ATCTCGTATGCCGTCTTCTGCTTG

[0029] The next core step after library preparation is cluster generation, which generates a sequence of amplified single template polynucleotides on a solid support. As explained above, to allow amplification of the template polynucleotides to proceed, a mixture of at least two amplification primers is immobilized, or "grafted," onto the surface of a suitable solid support.

[0030] As used herein, the term "solid support" refers to a rigid substrate that is insoluble in aqueous liquids (or at least rigid enough that it does not substantially swell when the liquid is removed and does not substantially shrink when the liquid is removed by drying). The present invention may utilize solid supports comprised of a substrate or matrix (e.g., glass slides, polymer beads, etc.) that has been "functionalized" by the application of a layer or coating of an intermediate material that contains reactive groups that allow for covalent attachment to biomolecules such as polynucleotides. A solid support may be a planar substrate. A solid support may be a layered substrate. A solid support may be a bead. Further examples of structured substrates or materials that may be used in the methods of the present disclosure are described in U.S. Patent Application Publication No. 13 / 661,524 and U.S. Patent Application Publication No. 2012 / 0316086 A1, which are incorporated herein by reference.

[0031] As a first step in colony generation by solid phase amplification, a mixture of forward and reverse amplification primers may be immobilized or "grafted" onto the surface of a suitable solid support.

[0032] When referring to the immobilization or attachment of a molecule (e.g., a nucleic acid) to a solid support, the terms "immobilized" and "attached" are used interchangeably herein, and both terms are intended to encompass direct or indirect, covalent or non-covalent attachment, unless otherwise indicated explicitly or by context. While covalent attachment may be preferred in certain embodiments of the invention, what is generally required is that the molecule (e.g., a nucleic acid) remain immobilized or attached to the support under conditions under which the support is intended to be used, for example, in applications requiring nucleic acid amplification and / or nucleic acid sequencing. When referring to the attachment of a nucleic acid to another nucleic acid, then the terms "immobilized" and "hybridized" are used herein and generally refer to hydrogen bonding between complementary nucleic acids.

[0033] In one embodiment, the grafting step generally involves covalent attachment of the primer to the support at or near the 5' end, with no primer extension at the 3' end.

[0034] Amplification primers are typically oligonucleotide molecules having the following structure: Forward primer: AL-S1 Reverse primer: AL-S2 where A represents an optional moiety that allows for attachment to a solid support, L represents an optional linker moiety, and S1 and S2 are polynucleotide sequences that allow for amplification of a substrate nucleic acid molecule that includes (fully or partially) a target region whose sequence is desired. The sequences S1 and S2 in the amplification primers may be specific to the particular target nucleic acid that is desired to be amplified, although in other embodiments, the sequences S1 and S2 may be "universal" primer sequences that allow for the amplification of any target nucleic acid of known or unknown sequence that has been modified (e.g., with an adapter as described above) to allow amplification with a universal primer.

[0035] The mixture of primers grafted onto the solid support generally contains substantially equal amounts of forward and reverse primers.

[0036] Group A can be any moiety (including non-nucleotide chemical modifications) that allows for attachment (preferably covalent attachment) to a solid support. Group A may include a sulfur-containing nucleophile, such as a phosphorothioate, present at the 5' end of a polynucleotide chain. Alternatively, Group A may be omitted if suitable chemistries are used to directly attach either the linker or the nucleic acid to the solid support.

[0037] L represents a linker or spacer that may be included, but is not strictly necessary. The linker may be included to ensure that the cleavage site present in the immobilized polynucleotide molecule produced as a result of the amplification reaction is positioned at an optimal distance from the solid support, or the linker itself may contain the cleavage site.

[0038] The linker has the formula (CH2) n where "n" is from 1 to about 1500, for example, less than about 1000, preferably less than 100, for example, from 2 to 50, and particularly from 5 to 25. However, a variety of other linkers may be used, with only restrictions placed on their structure such that the linker is stable under the conditions under which the polynucleotide is intended to be subsequently used, for example, under the conditions used in DNA amplification or DNA sequencing.

[0039] Linkers that do not consist solely of carbon atoms may also be used. Such linkers may include polyethylene glycol (PEG).

[0040] Linkers formed primarily from a chain of carbon atoms and PEG may be modified to include chain-interrupting functional groups. Examples of such groups include ketones, esters, amines, amides, ethers, thioethers, sulfoxides, and sulfones. Separately or in combination with the presence of such functional groups, alkenes, alkynes, aromatic or heteroaromatic moieties, or cycloaliphatic moieties (e.g., cyclohexyl) may also be used. Cyclohexyl or phenyl rings may be linked, for example, via their 1- and 4-positions, to PEG or (CH2) n They may be connected in a chain.

[0041] As an alternative to the above linkers based on a linear chain of predominantly saturated carbon atoms, optionally interrupted by unsaturated carbon atoms or heteroatoms, other linkers based on nucleic acids or monosaccharide units (e.g., glucose) may be envisaged. It is also within the scope of the present invention to utilize peptides as linkers.

[0042] In further embodiments, the linker may comprise one or more nucleotides. Such nucleotides may also be referred to herein as "spacer" nucleotides. Typically, 1 to 20, more preferably 1 to 15, or 1 to 10, more specifically 2, 3, 4, 5, 6, 7, 8, 9, or 10 spacer nucleotides may be included. Most preferably, the primer comprises 10 spacer nucleotides. It is preferred to use a poly-T spacer, although other nucleotides and combinations thereof may be used. In a preferred embodiment, the primer may comprise 10T spacer nucleotides.

[0043] For the primer grafting reaction to proceed, a mixture of amplification primers is applied to the solid support under conditions that allow reaction between moiety A (if present) and the support, or between the nucleic acid and the carrier. The solid support may be suitably functionalized to allow covalent attachment via moiety A. The result of the grafting reaction is a substantially uniform distribution of primers across at least a portion of the solid support. If the solid support comprises nanowells, then in a preferred embodiment, the primers are restricted to the locations of the nanowells and are absent from the interstitial regions of the solid support.

[0044] Following binding of the amplification primers, the solid support is contacted with a template polynucleotide and amplified under conditions that allow hybridization between the template and the immobilized primers. In the above example, the P5' and P7' primer binding sequences on the template polynucleotide are complementary to the immobilized primers (P5 and P7, respectively) present on the surface of the solid support. As used herein, "'" indicates the complementary strand.

[0045] The template is usually added in free solution under suitable hybridization conditions, as will be apparent to those skilled in the art. Typically, hybridization conditions are, for example, 5xSSC at 40°C. Solid-phase amplification can then proceed (e.g., by methods similar to those of WO 98 / 44151) with the first step of amplification being a primer extension step, in which nucleotides are added to the 3' end of the immobilized primer hybridized to the template to generate a fully extended immobilized complementary strand (or polynucleotide duplex). This complementary strand thus contains a sequence at its 3' end that can bind to or "crosslink" to a second primer molecule immobilized on the solid support, resulting in a new round of amplification beginning with the extension of the second immobilized primer using the complementary strand as a template.

[0046] Subsequent amplification reactions may then proceed substantially as described in WO 98 / 44151 or WO 00 / 18957, resulting in the generation of clustered arrays comprised of colonies of "bridged" amplification products. Here, both strands of the amplification product may be immobilized on a solid support at or near their 5' ends, with this attachment derived from the original attachment of the amplification primer. Typically, the amplification products within each colony are derived from the amplification of a single template (target) molecule. Alternatively, other amplification procedures may be used to generate clusters, as would be known to those skilled in the art. For example, amplification may be isothermal amplification using a strand-displacing polymerase, or may be exclusive amplification as described in WO 2013 / 188582.

[0047] Modifications necessary to allow subsequent cleavage of the crosslinked amplification product may be advantageously included in one or both amplification primers. Such modifications may be placed anywhere in the amplification primer and do not affect the efficiency of the amplification reaction to any extent. Thus, the cleavage-enabling modification may form part of the linker region L, or one or both of sequences S1 or S2. By way of example, amplification primers may be modified to contain, among other things, diol linkages, uracil nucleotides, ribonucleotides, methylated nucleotides, peptide linkers, PCR stoppers, or recognition sequences for restriction endonucleases. Because all nucleic acid molecules prepared by solid-phase amplification may ultimately contain sequences derived from the amplification primers, any modifications in the primers may be carried over into the amplification product.

[0048] In this context, the term "solid-phase amplification" refers to an amplification reaction that is similar to standard PCR, except that the forward amplification primer and / or the reverse amplification primer are immobilized (e.g., covalently attached) to a solid support at or near their 5' ends. Thus, the products of the PCR reaction are extended strands derived by extension of the amplification primers that are immobilized on a solid support at or near their 5' ends. Solid-phase amplification may be carried out using procedures similar to those described, for example, in WO 98 / 44151 and WO 00 / 18957.

[0049] A polynucleotide duplex is typically formed from two complementary polynucleotide strands composed of deoxyribonucleotides linked by phosphodiester bonds, but may also contain one or more ribonucleotides and / or non-nucleotide chemical moieties and / or non-naturally occurring nucleotides and / or non-naturally occurring backbone linkages. In particular, double-stranded nucleic acids may contain non-nucleotide chemical moieties, such as linkers or spacers, at the 5'-end of one or both strands. By way of non-limiting example, double-stranded nucleic acids may contain methylated nucleotides, uracil bases, phosphorothioate groups, ribonucleotides, diol bonds, disulfide bonds, peptides, etc. Such non-DNA or non-natural modifications may be included to enable cleavage or to confer some other suitable property, such as to enable covalent attachment to a solid support or to act as a spacer to position the cleavage site at an optimal distance from the solid support.

[0050] When a polynucleotide strand is only partially hybridized to a complementary strand, for example, a long polynucleotide strand hybridized to a short nucleotide primer, it may be referred to herein as a single-stranded nucleic acid.

[0051] The next core step is to sequence the cluster sequence.

[0052] To facilitate sequencing, it is preferred that one of the strands be removed from the surface, e.g., the second strand be removed leaving the first strand, or vice versa, to allow efficient hybridization of a sequencing primer to the remaining immobilized strand. Suitable methods for linearization are described below and in more detail in WO 07010251, the contents of which are incorporated herein by reference in their entirety.

[0053] Denaturation (and subsequent reannealing of the cleaved strand) results in the generation of a sequencing template that is partially or substantially single-stranded. The sequencing reaction may then be initiated by hybridization of a sequencing primer to the single-stranded portion of the template. In embodiments of the invention, sequencing may be carried out using a strand-displacing polymerase enzyme, as described below.

[0054] This method may require performing two sequencing reactions. The first sequencing reaction may be initiated by a first sequencing primer added from solution, resulting in sequencing of the first template strand, or by the 3' hydroxyl group of the immobilized primer being released from the immobilized duplex. The second sequencing reaction may be initiated by a second sequencing primer, which may be either immobilized or applied in solution. Hybridization of the in-solution sequencing primer to the (first) template strand is achieved by contacting the primer and template strand under conditions that promote annealing of the primer to the template. Such conditions are generally well known to those skilled in the art of molecular biology. Nevertheless, this method allows for obtaining sequence data from both ends of the template polynucleotide by obtaining a sequence read from one strand of the template, copying the strand using an immobilized primer (as described below), terminating the first strand, and sequencing the second copied strand. This provides sequence reads from both ends of the original fragment.

[0055] Thus, in one embodiment, the present invention relates to a method for sequencing regions of a polynucleotide template, referred to herein as a first region and a second region for sequencing, where the first and second regions for sequencing are at opposite ends of the polynucleotide template (the template and its complement are referred to herein as the first template strand and the second template strand, respectively). Because the sequence of a strand is known, the sequence of its complementary strand is also known; thus, the term two regions can apply equally to opposite ends of a single-stranded template or opposite ends of a double-stranded template, where the first region and its complement are known and the second region and its complement are known.

[0056] In one example, the first sequencing read may include binding of a first sequencing primer (read 1 sequencing primer) to a first sequencing binding site (e.g., SBS3'), followed by synthesis and sequencing of a complementary strand. This results in sequencing of the insert. In a second step, an index sequencing primer (e.g., i7 sequencing primer) binds to a second sequencing binding site (e.g., SBS12), resulting in synthesis and sequencing of an index sequence (e.g., sequencing of the i7 primer). Following cluster renaturation (as described herein), there is a second sequencing read. The second sequencing read may include binding of an index sequencing primer (e.g., i5 sequencing primer) to the complement of the first sequencing binding site (e.g., SBS3) on the template, and synthesis and sequencing of an index sequence (e.g., i5). In the second step, a second sequencing primer (Read 2 sequencing primer) binds to the complement of the primer (e.g., i7 sequencing primer) and binds to a second sequencing binding site (e.g., SBS12') to result in synthesis and sequencing of the insert in the reverse direction. In other words, SBS3 is the sequencing primer for Read 1, SBS12 for Read 2, SBS3' for i5, and SBS12' for i7.

[0057] Sequencing can be performed using any suitable "sequencing-by-synthesis" technique, in which nucleotides or oligonucleotides are subsequently added to the free 3' hydroxyl group, typically provided by annealing of a sequencing primer, resulting in the synthesis of a polynucleotide chain in the 5' to 3' direction. In certain embodiments, the nature of the nucleotide or oligonucleotide added is determined after each addition.

[0058] One particular sequencing method relies on the use of modified nucleotides that can act as reversible chain terminators. Suitable nucleotides are described in International Publication No. WO 04018497. When a modified nucleotide is incorporated into a growing polynucleotide strand complementary to the region of the template being sequenced, no free 3'-OH group is available to guide further sequence extension, and therefore the polymerase cannot add additional nucleotides. Once the nature of the base incorporated into the growing strand is determined, the 3' block can be removed to allow the addition of the next successive nucleotide. By sequencing the products derived using these modified nucleotides, it is possible to deduce the DNA sequence of the DNA template. Such reactions can be performed in a single experiment if each modified nucleotide is attached to a different label known to correspond to a specific base, facilitating discrimination between the bases added at each incorporation step. Suitable labels are described in PCT Application No. PCT / GB / 2007 / 001770, the contents of which are incorporated herein by reference in their entirety. Alternatively, separate reactions containing each separately added modified nucleotide can be performed.

[0059] Modified nucleotides may also be labeled to facilitate their detection. In certain embodiments, the label is a fluorescent label. Each type of nucleotide may be labeled with a different fluorescent label. However, the detectable label does not have to be a fluorescent label. Any label that allows the detection of the incorporation of nucleotide into DNA sequence may be used.

[0060] One method for detecting fluorescently labeled nucleotides involves the use of laser light of a wavelength specific to the labeled nucleotide, or other suitable illumination source. Fluorescence from the label on the incorporated nucleotide may be detected by a CCD camera or other suitable detection means. Suitable detection means are described in PCT / US2007 / 007991, the contents of which are incorporated herein by reference in their entirety. Once the first sequencing read is completed and a sufficient read length has been determined, the remaining strand can be copied. If a 3'-hydroxyl group was initially created with a cleavage enzyme, a new 3'-hydroxyl group can then be recreated at the same position, and extension can occur from this position, but it is also possible to continue copying the first template strand from the 3'-hydroxyl group of the nucleotide incorporated as part of the sequencing reaction. This extension reaction with all four unlabeled nucleotides and a polymerase can copy all bases of the first template. The immobilized primer may contain a restriction site for a cleaving enzyme, and treatment with the restriction enzyme may shorten the immobilized primer or immobilized template duplex, releasing an unblocked 3' hydroxyl group.

[0061] Methods for generating a free 3'-hydroxyl on only one strand of a duplex include either treatment with a cleavage enzyme or chemical treatment to remove specific nucleotides. Suitable cleavage enzymes are well known in the art, and preferably cleave at a site 3' distant from their binding site to avoid sequencing bases derived from known nick sites. The cleavage enzyme needs to cleave only one of the strands at the end closest to the surface. Examples of suitable restriction enzymes include Nt.BstNBI and Nt.AIwI, which do not have a defined sequence of bases beyond the released 3'-hydroxyl.

[0062] After the first sequencing run, the nucleotides undergo a regeneration step (as discussed further herein) to allow for a second sequencing run. Thus, the nucleotides may be deprotected to allow for further cycles of copying the template strand. If the nucleotides carry a dideoxy modification, this can be removed using an exonuclease or a polymerase with exonuclease activity. Thus, a cluster can be created with two grafted primers as described, with the unused primer blocked during the first sequencing reaction and then deblocked to allow for further copying of the template strand. Some of the amplification primers may be attached to the surface with a modification that blocks the 3' hydroxyl from extension during the amplification cycle. This means that the surface is treated with three or more amplification primers instead of two. At least two of the amplification primers should contain regions of identical sequence, but at least one primer may contain a 3'-blocking portion that is not affected by the conditions used to remove the second primer during the linearization process. Blocking moieties may take the form of chemical blocks, such as azidomethyl groups that can be removed with phosphine reagents, enzymatically removable, such as phosphate groups that can be removed with phosphatases, or nucleoside groups that can be removed using 3'-5' exonucleolysis. Such nucleoside modifications include abasic sites, which can be removed as described, or 2',3' dideoxynucleotides, which can be removed by polymerases with exonuclease activity. Further modifications include using oligonucleotide sequences that can form self-complementary regions with restriction enzyme recognition sequences. Treatment with the restriction enzyme is required to cleave the hairpin strand and release a shorter sequence with a free 3' hydroxyl group. Treatment of the surface after the first sequencing run can deblock the primer, allowing the remaining first strand to hybridize to the deprotected primer and recopy the previously sequenced strand.

[0063] Once the sequencing product of the first sequencing read is removed, the first template strand may remain immobilized on the solid support. The first template strand contains, at its 3' end, a sequence complementary to a second primer immobilized on the solid support, which can bind or "bridge" to that second primer molecule. An extension reaction may then be performed as further described herein, extending one or more of the immobilized primers to copy the first template strand and generate a second immobilized template strand, which is then sequenced with the second sequence read. Thus, the cluster may be regenerated to allow for a second sequence read. This process may be repeated multiple times to obtain paired-end resynthesis. Figure 7 shows an example of this cycle. Here, bridge amplification requires three cycles of reagents to perform a multiplication round. A first denaturing agent (LDR) is used to denature the cluster at 55°C. Next, HO is pumped across the solid support (e.g., a flow cell) to facilitate removal of the denaturing agent. A strand-displacing polymerase enzyme is then used to extend the available 3' surface primer, and the cycle may be repeated as described below for paired-end resynthesis.

[0064] Again, this recreates the template duplex in which both strands are immobilized. For the extension reaction during pairwise sequencing (i.e., between the first and second sequencing reads), the extension is carried out using a strand-displacing polymerase enzyme.

[0065] In a further step, it may be advantageous to extend the free 3'-hydroxyl primer with multiple bases complementary to the template before initiating sequencing. Both of these actions increase the melting temperature of the immobilized duplex and help prevent the template strand from rehybridizing to other immobilized primers during sequencing, but this can lead to cascading intracluster problems. The ligation step is performed after the phosphatase step removes the phosphate group from the immobilized primer. Addition of 20-30 bases of sequence can be achieved by ligation with a 5'-phosphate-modified primer hybridized adjacent to the free 3' hydroxyl. A ligase such as T4 DNA ligase can be used to seal the gap. In the case of a USER treatment to remove U nucleotides, the 5'-base of the primer can be a T to replace the excised U. For the hybridization step to be performed efficiently, the 5'-non-immobilized strand must have been removed by the 5'-3' exonucleolysis treatment described above. Such immobilized extended primers with a free 3'-hydroxyl are described as extended 5' anchors or extended immobilized primers, although the generation of such extended primers is only one of the steps involved in processing multiple template polynucleotides such that a first template strand is hybridized at its 5' end to a primer that is immobilized on a solid support.

[0066] In one example, it may be advantageous to extend the immobilized primer before carrying out the extension reaction described above. Extension may be performed using a hybridized oligonucleotide having a sequence that extends beyond the 3' end of the immobilized primer, which sequence is also the same as the corresponding region at the end of the template. This extension may serve as a base for extension of the immobilized primer, and thus the extended primer is complementary to the immobilized template strand. Due to their increased length, extended primers may improve the efficiency of the strand resynthesis process. Once the complementary sequence of the first strand is generated, the first strand may be removed from the surface.

[0067] Removal of the first template from the surface allows a new single-stranded second template to be sequenced, again starting from the 3' end. Thus, both ends of the original immobilized template can be sequenced. The first strand can be removed by a suitable orthogonal linearization process, such as diol cleavage or removal of the 8-oxo-G residue. After denaturation of the first template strand, a second sequencing primer can be hybridized to the second template strand, allowing sequencing of the second template strand. This orthogonal linearization strategy also allows for reads from both ends of the template.

[0068] Selective removal or linearization of the first template strand can also be achieved in several other ways. Linearization, which allows for hybridization of a sequencing primer in solution, does not necessarily leave a functional 3'-hydroxyl on the template strand and can cleave either one or both strands. Thus, as used herein, the term "linearization" refers to the selective removal of the complementary strand. If one of the amplification primers is immobilized so that it can be cleaved from a surface, the resulting double-stranded DNA can be rendered single-stranded using heat or chemical denaturing conditions to give a single-stranded molecule containing a primer hybridization site. The single-stranded molecule can be hybridized with a sequencing primer in solution to allow sequencing reads of the immobilized template strand. The cleavage site allows for controlled cleavage of the first template strand by chemical, enzymatic, or photochemical means. Any suitable enzymatic, chemical, or photochemical cleavage reaction may be used for cleavage. Many suitable methods are described in International Publication No. 07010251, the contents of which are incorporated herein by reference in their entirety. The cleavage reaction may result in partial or complete removal of the cleaved strand. Suitable cleavage methods include, for example, restriction enzyme digestion, where the cleavage site is a suitable restriction site for the enzyme that induces cleavage of one or both strands of the double template, RNase digestion or chemical cleavage of bonds between deoxyribonucleotides and ribonucleotides, where the cleavage site may include one or more ribonucleotides, chemical reduction of disulfide bonds with a reducing agent (e.g., TCEP), where the cleavage site should include an appropriate disulfide bond, chemical cleavage of diol bonds with periodate, where the cleavage site should include a diol bond, generation of an abasic site followed by hydrolysis, etc.

[0069] In one embodiment, cleavage may occur at the cleavage site on one or both strands of a template polynucleotide duplex that includes one or more or any combination of non-natural nucleotides, ribonucleotides, or non-nucleotide chemical modifications.

[0070] Suitable cleavage techniques for use in the methods of the present invention include, but are not limited to, the following, which are described in detail in WO2008041002: i) chemical cleavage, ii) abasic site cleavage, iii) ribonucleotide cleavage, iv) photochemical cleavage, v) PCR stopper, vi) peptide linker, vii) enzymatic digestion.

[0071] A second sequencing primer is then hybridized to the copied strand of the template, and the sequencing reaction proceeds as described above through the sequential addition of nucleotides to the second sequencing primer, resulting in the determination of the sequence of a second region of the target polynucleotide.

[0072] The polynucleotide duplexes described herein form part of a single cluster or colony composed of many such first and second duplexes, and the cluster or colony itself may typically form part of an array of many such clusters or colonies. The terms "cluster" and "colony" are used interchangeably throughout to refer to a distinct site on a solid support consisting of a plurality of identical immobilized nucleic acid strands and a plurality of identical immobilized complementary nucleic acid strands. The term "clustered array" refers to an array formed from such a cluster or colony. Each polynucleotide duplex in the array contains the same universal primer recognition region, allowing the same primer to be used for each cluster sequence. As described above, a first sequencing primer is then hybridized to the first template strand, and the sequencing reaction proceeds through the sequential incorporation of nucleotides or oligonucleotides into the first sequencing primer, resulting in the determination of the sequence of a first region of the target polynucleotide.

[0073] An important feature is that both sequencing runs can occur in the same cluster or colony on the clustered array. The sequence of each duplex within each such colony contains the same double-stranded target polynucleotide, but different colonies may be formed from duplexes containing different double-stranded target polynucleotides. In certain embodiments, at least 90%, more particularly at least 95%, of the colonies on a given clustered array can be formed from template duplexes containing different double-stranded target polynucleotides, but within each individual colony on the array, all template duplexes can contain the same double-stranded target polynucleotide.

[0074] The sequencing methods outlined above are not limiting, and essentially any sequencing methodology that relies on the sequential incorporation of nucleotides into a polynucleotide chain may be used. Suitable techniques include, for example, Pyrosequencing™, FISSEQ (fluorescent in situ sequencing), MPSS (massively parallel sequencing), and ligation-based sequencing. The target double-stranded polynucleotide to be sequenced may be any polynucleotide for which sequencing is desired. The target polynucleotide may have a known, unknown, or partially known sequence, such as for resequencing applications. Using the template preparation method described in detail below, it is possible to prepare a template sequence starting from essentially any double-stranded target polynucleotide of known, unknown, or partially known sequence. The use of sequences allows for parallel sequencing of multiple targets of the same or different sequences. A particular application of pairwise methods is in sequencing fragments of genomic DNA. This method offers particular advantages in identifying genomic rearrangements, as the two regions of sequence obtained for each target molecule using this method are known to be linked within a specific distance of each other in the genome, depending on the size of the starting target molecule.

[0075] The present invention identifies that a non-thermostable strand-displacing polymerase can be advantageously used during the regeneration step after the first sequencing read and before the second sequencing read. In other words, the non-thermostable strand-displacing polymerase can be used to resynthesize the complementary strand of the first template strand after sequencing of that strand. As explained above, this is called paired-end resynthesis.

[0076] "Non-thermostable polymerase" is intended to encompass polymerases that have been optimized to operate at lower temperatures relative to thermostable polymerases such as BST.

[0077] A "strand-displacing polymerase" exhibits the ability to displace downstream DNA that it encounters during synthesis.

[0078] In one embodiment, the non-thermostable polymerase according to the invention is capable of heating at temperatures below 55°C. Optimal In further embodiments, the non-thermostable polymerase according to the invention has an incubation temperature of less than 50°C, less than 45°C, less than 40°C, less than 39°C, 38°C or less, about 38°C, or about 37°C. Optimal Particularly preferred non-thermostable polymerases have an incubation temperature of about 38°C. Optimal It has an incubation temperature.

[0079] In a further embodiment, the non-thermostable polymerase according to the invention is capable of heating at temperatures below 55°C. Optimal In further embodiments, non-thermostable polymerases according to the invention have an activity temperature of less than 50°C, less than 45°C, less than 40°C, less than 39°C, 38°C or less, about 38°C, or about 37°C. Optimal Particularly preferred non-thermostable polymerases have an activity temperature of about 38°C. Optimal It has an activation temperature.

[0080] Suitable non-thermostable strand-displacing polymerases according to the present invention can be found, for example, by New England BioLabs, Inc. and include phi29, Bsu, Klenow, and DNA Polymerase I (E. coli) and functional fragments thereof. A particularly preferred polymerase is Bsu. In an alternative embodiment, the polymerase is the Bsu large fragment.

[0081] By using a non-thermostable polymerase, the extension reaction can be carried out at a temperature below 55°C, preferably below 50°C, preferably below 40°C, preferably 38°C + / - 2°C, preferably 38°C + / - 1°C, preferably 38°C.

[0082] The use of a strand-displacing polymerase can also provide advantages and overcome challenges due to the reannealing of the template strand to itself. The use of a strand-displacing polymerase helps the template strand anneal to the associated primer, allowing for efficient extension.

[0083] The use of a non-thermostable polymerase during the regeneration step after the first sequencing read offers numerous advantages. For example, the use of thermostable BST has previously been used during this regeneration step, which requires 12 regeneration cycles to achieve optimal % resynthesis. In contrast, non-thermostable polymerases have been shown to achieve acceptable resynthesis after only three cycles. See, for example, the PETv23x5m protocol in the Examples, which achieved 79.4% resynthesis after only three cycles.

[0084] The use of a non-thermostable polymerase during paired-end resynthesis is counterintuitive, as this step is between two sequencing steps, both of which are performed at high temperatures using a suitable SBS polymerase that can operate at 60 or 65°C.

[0085] The use of a non-thermostable polymerase during paired-end resynthesis may also provide additional benefits if the same polymerase is used during the initial cluster amplification to generate cluster sequences that are subsequently sequenced. For example, if ExAmp (an amplification mixture containing the non-thermostable strand-displacing polymerase BSU) is used in both the initial cluster generation step and the paired-end resynthesis step, then the total number of reagents used in the machine is reduced, and the cartridge complexity is similarly reduced. Such benefits may result in COGS optimization and protocol simplification.

[0086] In one embodiment, step (d) is repeated with multiple cycles of extension and denaturation, which may include 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more cycles, preferably 2-5 cycles, 2-4 cycles, 3-5 cycles, 3 cycles, or 4 cycles, most preferably 3 cycles. In contrast to polymerases that require a greater number of cycles, the present invention allows for efficient completion of the renaturation step.

[0087] In one embodiment, each cycle may last 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 20, or 30 minutes. A particularly preferred cycle type is 5 minutes. In certain embodiments, step (d) may include the following extension and denaturation cycles: 3 x 5 minutes; 3 x 30 minutes; 12 x 2 minutes; 6 x 4 minutes; 2 x 12 minutes, or 3 x 5 minutes. In certain embodiments, 3 x 5 minutes is preferred.

[0088] In a preferred embodiment, the extension reaction during pairwise sequencing comprises cycles of extension and denaturation.

[0089] Any suitable denaturant is encompassed by the present invention. Suitable denaturants include: acidic nucleic acid denaturants such as acetic acid, HCl, or nitric acid; basic nucleic acid denaturants such as NaOH; or other nucleic acid denaturants such as DMSO, formamide, betaine, guanidine, sodium salicylate, propylene glycol, or urea. Preferred denaturants are formamide and NaOH, preferably formamide.

[0090] In one embodiment, denaturation is carried out at substantially the temperature of the extension reaction.

[0091] In a particularly preferred embodiment, the extension / denaturation cycle comprises three cycles of extension with two cycles of denaturation between them. In a further preferred embodiment, a denaturation cycle is performed before the first extension cycle. Such an embodiment comprises three alternating cycles of denaturation and extension. Formamide is the most preferred denaturant.

[0092] In one embodiment, the extension reaction (d) is carried out at a lower temperature than the sequencing read step, which may generally use a different polymerase (e.g., SBS polymerase) optimized for higher temperatures.

[0093] In one embodiment, the plurality of double-stranded template polynucleotides is generated by an amplification reaction using the same polymerase as used in the extension reaction step (d).

[0094] The plurality of double-stranded template polynucleotides immobilized on the solid support in step (a) may preferably be a cluster. While the present invention encompasses any method for generating the cluster, one preferred methodology uses bridge amplification. In a preferred embodiment, the bridge amplification to generate the cluster uses the same non-thermostable strand-displacing polymerase used in the extension reaction of step (d). Such an approach can minimize the amount of amplification mixture used throughout the process, thereby streamlining further methodology.

[0095] Therefore, a single non-thermostable strand-displacing polymerase can be used for all amplification steps to generate or regenerate immobilized template polynucleotides. A second polymerase can be used during the sequencing step, meaning that the entire process can be completed with just two polymerases. This streamlined approach offers various process advantages, reduced complexity, and cost savings.

[0096] The terms "cluster" and "colony" are used interchangeably throughout and refer to distinct sites on a solid support consisting of a plurality of identical immobilized nucleic acid strands and a plurality of identical immobilized complementary nucleic acid strands. The term "clustered array" refers to an array formed from such a cluster or colony.

[0097] The present invention provides a method for pairwise sequencing of a first region and a second region of a target double-stranded polynucleotide, wherein the first region and the second region are in the same target double-stranded polynucleotide, and the pairwise sequencing comprises steps (a) to (f) disclosed herein.

[0098] The method may be used to obtain two linked or paired reads of sequencing information from each of the double-stranded templates on the clustered sequence.

[0099] The present invention provides a method for improving the data quality of a sequencing reaction, comprising performing an extension reaction as disclosed herein.

[0100] The invention will now be illustrated by the following non-limiting examples. [Example]

[0101] Paired-end sequencing may be performed according to the method described in WO 2008 / 041002 (see, e.g., Example 13), with the following modifications described below for the first sequencing step of the target fragment: Following successful completion of Read 1 sequencing, 1. Deprotection of the flow cell surface-bound i5 primer is carried out enzymatically using a resynthesis mix containing a buffer such as JRM available from Illumina, Inc. and T4 polynucleotide kinase. 2. The flow cell is then washed with an ionic buffer, for example, BB6 buffer available from Illumina Inc. 3. The cluster is denatured with low bias denaturing reagent (LDR-100% formamide) to remove the extended sequencing primer from Read 1. 4. Set the flow cell temperature to 38°C. 5. To regenerate the double-stranded clusters (resynthesis), three subsequent rounds of LDR (100% formamide) and ExAmp (an amplification mix containing the non-thermostable strand-displacing polymerase BSU) are cycled isothermally at 38°C as follows: LDR and ExAmp are available from Illumina, Inc. 5a. The LDR is flushed through the flow cell with a large flush factor intended to completely replace the contents of the flow cell. 5b. Next, flush ExAmp through the flow cell, maintaining a large flush factor, and incubate for 5 minutes. 5c. Repeat steps 5a and 5b for a total of three rounds of amplification. 6. The new double-stranded cluster is then linearized for read 2 and primer hybridization is performed as in WO 2008 / 041002.

[0102] A graphical representation of the workflow according to the present invention is shown in Figure 1. Workflow PETv2 shows deprotection with JRM followed by buffer washes. The clusters are then denatured with LDR, followed by thermocycling to 38°C. This is followed by three regeneration and denaturation rounds using ExAmp (containing BSU) and LDR (100% formamide).

[0103] An alternative workflow, PETv1, is shown in Figure 2 and is similar to the workflow in Figure 1, but does not include LDR denaturation between regenerations, instead proceeding with three consecutive regeneration rounds. Figures 3 and 4 mirror Figures 1 and 2, but also include linearization.

[0104] In embodiments, SSCs may also be used in protocols according to the invention, for example, clusters may be washed before and / or after recombination with SSCs.

[0105] The regeneration efficiency of these workflows was evaluated against other potential workflows, and the results are shown in Figure 5, which shows the % resynthesis calculated from HSX sequencing runs with the paired-end turn protocol according to the present invention, as described in Table 1 below. [Table 1]

[0106] The initial protocol tested was PETv1 after 3 × 5 min cycles. This achieved 64% resynthesis, which is considered acceptable but not optimal. Increasing the incubation time to 3 × 30 min increased the resynthesis % to 84%, and increasing both incubation time and number of chemical cycles also had a positive impact on the resynthesis % (PETv2 12 cycles - 2 min, no LDR improved the resynthesis % to 71%).

[0107] In the examples using LDR, the use of continuous renaturation / denaturation cycles further improved the % resynthesis to 76.5-101.2% depending on the conditions.

[0108] Increasing the ExAmp incubation time, number of chemical cycles, and / or frequency of LDR use can be seen to have a positive impact on % resynthesis. Typically, a selected protocol can achieve ∼80% resynthesis or more, but a preferred protocol may trade % resynthesis versus time versus protocol complexity and reagent use.

[0109] Protocol optimization was also considered using DOE on the iCbot system, the results of which are shown in Figures 6A and 6B.

[0110] Figure 6A shows that: i) LDR cycles have a significant effect on % resynthesis; i) increasing the number of cycle steps also improves % resynthesis, especially when LDR cycles are used; and iii) increasing incubation time has a positive but not significant effect on % resynthesis, suggesting that shorter run times are acceptable.

[0111] Figure 6B shows a model fit to experiments identifying optimized parameters: LDR = Yes; 4 pushes; 8.7 min incubation time. As discussed herein, the final optimization protocol may consider not only % resynthesis, but also overall optimization in terms of speed, reagent use, and efficiency.

[0112] In conclusion, the present invention identifies that extension reactions during pairwise sequencing to regenerate templates during a sequencing run can be carried out using non-thermostable strand-displacing polymerases at temperatures below 55°C. This provides process advantages. Furthermore, if the same polymerase is used for both the extension reaction and the initial cluster generation as well, then the overall process can be simplified, reducing the number of reagents and protocol steps.

[0113] Embodiments of the present invention can now be described with reference to the following sections. 1. A method for performing an extension reaction during pairwise sequencing of a first region and a second region of a target double-stranded polynucleotide, wherein the first region and the second region are in the same target double-stranded polynucleotide, and the pairwise sequencing comprises the steps of: (a) providing a solid support having a plurality of double-stranded template polynucleotides immobilized thereon, each formed from complementary first and second template strands linked to the solid support at their 5' ends and a plurality of copies of one or more 5'-end-immobilized primers capable of hybridizing to the 3' end of the first template strand; (b) selectively removing a second template strand of a plurality of double-stranded template polynucleotides to allow hybridization to a 5'-terminal immobilized primer of a first template strand; (c) performing a first sequencing read to determine the sequence of a first region of the template polynucleotide by sequencing-by-synthesis or sequencing-by-ligation techniques; (d) performing an extension reaction to extend one or more of the immobilized primers to copy the first template strand to generate a second immobilized template strand; (e) selectively removing a first template strand of the plurality of double-stranded template polynucleotides to allow hybridization of a sequencing primer to the template strand generated in step (d); (f) performing a second sequencing read to determine the sequence of a second region of the template polynucleotide by a sequencing-by-synthesis technique or by a sequencing-by-ligation technique, wherein determining the sequences of the first and second regions of the target polynucleotide achieves pairwise sequencing of the first and second regions of the double-stranded target polynucleotide; A method wherein the extension reaction of step (d) is carried out using a non-thermostable strand-displacing polymerase at a temperature below 55°C. 2. The non-thermostable polymerase has a temperature of less than 55°C, preferably less than 50°C, less than 45°C, less than 40°C, less than 39°C, less than 38°C, about 38°C, or about 37°C, particularly preferably about 38°C. Optimal Incubation temperature and / or Optimal Item 1, wherein the method has an activation temperature. 3. The method according to any of items 1 to 2, wherein the extension reaction is carried out at a temperature below 55°C, preferably below 50°C, preferably below 40°C, preferably 38°C + / - 2°C, preferably 38°C + / - 1°C, preferably 38°C. 4. The method according to any one of items 1 to 3, wherein the non-thermostable polymerase is Bsu, phi29, Klenow, or DNA Polymerase I (E. coli), preferably Bsu. 5. The method according to any one of items 1 to 4, wherein step (d) is repeated by multiple cycles of extension and denaturation. 6. The method according to item 5, wherein step (d) is repeated by multiple cycles of extension and denaturation. 7. The method according to item 5 or 6, comprising 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 or more cycles, preferably 2 to 5 cycles, 2 to 4 cycles, 3 to 5 cycles, 3 cycles, or 4 cycles, most preferably 3 cycles. 8. The method of any one of items 5 to 7, wherein each cycle can last 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 20, or 30 minutes. 9. The method according to item 6, comprising the following cycles of extension and denaturation: 3 x 5 min; 3 x 30 min; 12 x 2 min; 6 x 4 min; 2 x 12 min or 3 x 5 min; preferably 12 x 2 min. 10. The method according to item 6, wherein denaturation is carried out substantially at the temperature of the extension reaction. 11. The method according to any one of items 5 to 9, wherein the denaturation is carried out using an acidic nucleic acid denaturant such as acetic acid, HCl, or nitric acid; a basic nucleic acid denaturant such as NaOH; or other nucleic acid denaturants such as DMSO, formamide, betaine, guanidine, sodium salicylate, propylene glycol, or urea; preferably formamide and NaOH, particularly preferably formamide. 12. The method according to any one of items 1 to 11, wherein the extension reaction (d) is carried out at a lower temperature than the sequencing read step. 13. The method according to any one of items 1 to 12, wherein the plurality of double-stranded template polynucleotides immobilized on the solid support in step (a) is a cluster. 14. The method of item 13, wherein the clusters are generated by bridge amplification. 15. The method according to item 14, wherein the same non-thermostable strand-displacing polymerase is used during the initial cluster generation and extension reaction step (d). 16. The method according to any one of items 1 to 15, wherein the plurality of double-stranded template polynucleotides is generated by an amplification reaction using the same polymerase as that used in the extension reaction step (d). 17. The method according to any of items 1 to 16, wherein at least one of the immobilized primers is blocked at the 3' end, and the block is removed before step (d). 18. The method according to item 17, wherein the blocks are phosphate groups and the surface is treated with phosphatase to remove the blocks. 19. The method according to any of items 1 to 18, further comprising the step of treating with a restriction enzyme prior to step (d) to shorten the immobilized primer and release a free 3' hydroxyl for extension. 20. The method according to any of items 1 to 19, wherein the immobilized primer is extended before step (d). 21. The method according to item 20, wherein the immobilized primer is extended by hybridization of a non-immobilized complementary sequence with the 5'-overhang, and the immobilized primer is extended to copy the overhang. 22. A method for pairwise sequencing of a first region and a second region of a target double-stranded polynucleotide, wherein the first region and the second region are in the same target double-stranded polynucleotide, and the pairwise sequencing comprises steps (a) to (f) disclosed in any one of items 1 to 21. 23. Use of the method described in item 22 to obtain two linked or paired reads of sequencing information from each of the double-stranded templates on the clustered sequence. 24. A method for improving data quality of a sequencing reaction, comprising carrying out the extension reaction according to any one of items 1 to 21.

Claims

1. 1. A method for performing a strand resynthesis extension reaction during pairwise sequencing, wherein the strand resynthesis extension reaction is performed between a first sequencing read and a second sequencing read, the strand resynthesis extension reaction extending one or more immobilized primers to copy a first template strand to generate a second immobilized template strand, and the strand resynthesis extension reaction is performed using a non-thermostable strand displacement polymerase at a temperature below 55°C; the first template strand is a cluster, and the same non-thermostable strand displacement polymerase is used during the initial cluster generation and the strand resynthesis extension reaction.

2. 2. The method of claim 1, wherein the non-thermostable strand displacement polymerase has an optimum incubation temperature and / or optimum activity temperature of less than 55°C, less than 50°C, less than 45°C, less than 40°C, less than 39°C, 38°C or less, or 37°C.

3. The method of any one of claims 1 to 2, wherein the extension reaction is carried out at a temperature of less than 55°C, less than 50°C, less than 40°C, 38°C + / - 2°C, or 38°C + / - 1°C.

4. The method of any one of claims 1 to 3, wherein the non-thermostable strand displacement polymerase is Bsu, phi29, Klenow, and DNA polymerase I (E. coli), Bsu, or a functional fragment thereof.

5. The method of any one of claims 1 to 4, wherein the strand resynthesis extension reaction is repeated by multiple cycles of extension and denaturation.

6. 6. The method of claim 5, comprising 2 cycles, 3 cycles, 4 cycles, 5 cycles, 6 cycles, 7 cycles, 8 cycles, 9 cycles, 10 cycles, 11 cycles, 12 cycles, 13 cycles, 14 cycles, 15 cycles, or 16 or more cycles.

7. 7. The method of claim 5 or claim 6, wherein each cycle can last 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 20, or 30 minutes.

8. 8. The method of any one of claims 5 to 7, comprising the following cycles of extension and denaturation: 3 x 5 min; 3 x 30 min; 12 x 2 min; 6 x 4 min; 2 x 12 min or 3 x 5 min.

9. The method of any one of claims 5 to 8, wherein denaturation is carried out at substantially the same temperature as the temperature at which the extension reaction is carried out.

10. 10. The method according to any one of claims 5 to 9, wherein the denaturation is carried out using an acidic nucleic acid denaturant selected from acetic acid, HCl, and nitric acid; or a basic nucleic acid denaturant selected from NaOH, DMSO, formamide, betaine, guanidine, sodium salicylate, propylene glycol, and urea.

11. The method of any one of claims 1 to 10, wherein the strand resynthesis extension reaction is carried out at a lower temperature than the sequencing read step.

12. The method of claim 1 , wherein the clusters were initially generated by bridge amplification.

13. The method of any one of claims 1 to 12, wherein the first template strand is generated by an amplification reaction using the same polymerase as used in the strand resynthesis extension reaction.

14. The method of any one of claims 1 to 13, wherein at least one of the immobilized primers is blocked at the 3' end, and the block is removed prior to the strand resynthesis extension reaction.

15. 15. The method of claim 14, wherein the block is a phosphate group and is treated with a phosphatase to remove the block.

16. 16. The method of any of claims 1 to 15, further comprising the step of treating the immobilized primer with a restriction enzyme prior to the strand resynthesis extension reaction to shorten the immobilized primer and release a free 3' hydroxyl for extension.

17. The method of any one of claims 1 to 16, wherein the immobilized primer is extended prior to the strand resynthesis extension reaction.

18. 18. The method of claim 17, wherein the immobilized primer is extended by hybridization of a non-immobilized complementary sequence with a 5'-overhang, and the immobilized primer is extended to copy the overhang.

19. 19. A method for pairwise sequencing of a first region and a second region of a target double-stranded polynucleotide, wherein said first region and said second region are in the same target double-stranded polynucleotide, and said pairwise sequencing comprises the steps disclosed in any one of claims 1 to 18.

20. 20. Use of the method of claim 19 to obtain two linked or paired reads of sequencing information from each of the double-stranded templates on a clustered sequence.

21. A method for improving data quality of a sequencing reaction, comprising carrying out an extension reaction according to any one of claims 1 to 18.

Images