Multiple method for preparing sequencing library

Abstract

A method of preparing a library of library constructs for targeting next generation sequencing by multiplex amplification is described. The method comprises the steps of: (a) providing a reaction vessel comprising (i) a plurality of different target sequences, (ii) a plurality of target capture primer pairs, and (iii) one or more tagged primer pairs, (b) performing sequential rounds of amplification at sequential annealing temperatures configured to amplify the target sequences, producing a target sequence comprising first and second read sequences, and providing a reaction product comprising the library of constructs linked by the adapters in a sequential manner; and; and (c) capturing the library of constructs from the reaction product. One of the forward and reverse tagged primers comprises a purification label at the 5'end and is provided at a limited concentration, whereby the library construct comprises a multiplicity of partial constructs comprising only one index sequence and only one adapter sequence, and a limited number of (complete) intact constructs comprising first and second index sequences, first and second adapter sequences, and a purification tag. The capture step includes capturing a (fully) intact construct from the reaction product using a purification label.

Description

[0001] This application is a divisional application of patent application No. 202080064097.6, filed on September 11, 2020, entitled “A multiplexed method for preparing sequencing libraries”. TECHNICAL FIELD

[0002] The present invention relates to a multiplexed amplification method for preparing libraries for targeted next generation sequencing. Methods for targeted next generation sequencing comprising multiplexed amplification are also contemplated. BACKGROUND

[0003] Targeted genotyping by sequencing (GBS) involves the pooling of indexed and adaptor-ligated multiplexed amplification products into a single library sample for the run of high-throughput next generation sequencing. From this sequencing data, individual genotypic profiles are extracted and compiled by bioinformatics tools involving demultiplexing and allele counting. There are several modifications of this main method, open source and commercially sourced (Eureka and Agriseq Thermofisher, Nugen Allegro, Keygene SNPselect, LGC SeqSNP, AgriPlex Genomics PlexSeq). Open source methods generally rely on enzymatic amplification in enrichment and actual library preparation.

[0004] Developing an in-house workflow for targeted GBS methods can be divided into the following processes:

[0005] 1. Target-specific multiplexed amplification

[0006] 2. Sample indexing, sequencing primer and adaptor addition

[0007] 3. Normalization and size selection

[0008] 4. QC and final library pooling

[0009] 5. Next generation sequencing

[0010] 6. Data analysis

[0011] In all of the above methods, target amplification and library preparation are performed in multiple steps (steps 1 to 4). Having a high number of targets in the same amplification reaction can cause problems through non-specific binding and primer dimer formation. Therefore, to achieve a high performing assay panel, the multiplex nature of the initial reaction needs to be considered in the primer design step to avoid problematic sequences. The final panel will be fine-tuned through a small scale test iteration process. With amplification-based methods (GT-Seq, MonsterPlex, Hi-plex, PlexSeq), a multiplex level of up to about 500 is commonly reported. MonsterPlex (cited in the literature as similar to Hi-Plex) is expected to have about 85-90% of the markers working when run for the first time. These libraries are sequenced at high depth, about 500K reads per sample, to obtain good coverage. The Hi-Plex method, also based on amplification, claims an overlapping level of up to 1000 (Nguyen-Dumont et al. 2013, Nguyen-Dumont et al. 2015, Pope et al. 2018, Hammet et al. 2019). Hi-Plex is the only method that aims to combine two amplification steps; after the initial amplification, primers that introduce the index / adapters are incorporated into the multiplex reaction. Therefore, this method still requires opening and resealing of the sample plate. Key points listed by the authors in the principle process are high-fidelity Taq enzyme, allowing initial amplification conditions that allow simultaneous amplification of a series of targets, followed by a large amplification of the tail added for target-specific primers, and relatively long cycles that allow complete priming and extension. The method originally published for a 60-plex panel was further developed to achieve a 1003-plex by adding another set of so-called short bridge primers to the multiplex. The target-specific amplification mix here has two shorter primers, complementary to parts of the tail added for the target-specific primers. A large amplification is performed in the first part using the target-specific and bridge primers. The tagging primers are added to the mix until the last 4 cycles. This is different from the original 60-plex method, which does not use short bridge primers, where most of the amplification takes place after the addition of the index primers. The latest publication by Hammet et al. (2019) describes a further update to the Hi-Plex method to improve on-target amplification. In the updated method, the initial two-step amplification using target-specific and tagging primers has fewer cycles, followed by a size selection before performing another amplification to amplify the constructed library, but still requires opening and resealing of the sample plate. The universal short bridge primers are no longer used. The comparison by Hammet of the previous and current methods indicates similar performance at lower overlapping levels below 300, but the updated method improves capacity at higher numbers of targets.

[0012] Uniform annealing between the target primers is crucial for a more uniform amplification of multiple targets at the same time. The GT-Seq method does not perform longer cycles at a single temperature that is universal for all Hi-plex versions, but achieves this by including a slow temperature ramp from 95°C to 57°C to achieve more accurate annealing in 10 amplification cycles. Chen et al. (2016) introduce another method to balance amplification by using limited primer availability; the multiplex amplification products are used as targets in a second amplification step without the need to add more primers. This reaction is run to exhaustion of all available primers before indexing and results in a more normalized output (the authors call this method three-round amplification). The introduction of a third amplification reaction is not ideal, but verifying the correct primer availability in multiplex reactions can be helpful. Like Hi-Plex, the GT-Seq and MonsterPlex methods use high-quality Taq enzymes.

[0013] To pool samples into one sequencing run, each sample needs a unique label. This can be done by unique combinations of short sequences (e.g. 6- or 8-mers) on both ends or only one end of the DNA sample. To enable thousands of individuals to be sequenced in the same indexed run, a systematic indexing method is needed; like one index represents the well position and another one the plate. This high-density sample pool can be achieved by reusing 96- or 384-well specific indices in combination with different plate indices. In the amplification-based targeted GBS method, indexing is done by target-specific multiplex primers that carry a universal tail sequence; the forward primers share the same tail sequence, which is different from the tail shared by the reverse primers. These tails are targeted in a second amplification step by primers that add a unique index and a sequencing adapter. In addition, the universal tail introduced by the target-specific primers is used as sequencing primer target in the sequencing reaction. After the second amplification step, a sample-specific library is obtained that contains the multiplexed targets.

[0014] The process to link these two reactions can differ. Like already mentioned by Chen et al. (2016), an amplification step can be included between the two to exhaust all target-specific primers. In GT-Seq, the amplification products from the multiplex reaction are diluted and then used as targets in a new plate in an amplification that adds a second label / adapter. The Hi-Plex method uses a different approach. The indexing primers are added directly to the initial amplification plate without an intermediate dilution step. This reaction uses the enzyme mix that was added in the first amplification setup, a method that saves both reagents and plates. But this plate still needs to be pierced or peeled and further re-sealed. The last step is difficult to apply in high-throughput due to the risk of sample contamination, especially in case of water bath thermocycling.

[0015] GB2536446 describes a single closed tube PCR method for generating amplicon constructs of target sequences by reverse complement PCR (RC-PCR). The method uses (a) an oligonucleotide probe comprising a universal sequence and a target-specific sequence at or proximal to its 5' end, which is capable of hybridising to the reverse complement of a sequence located at one of the 3' ends of the target sequence or flanking it, and (b) a universal primer comprising at its 3' end a sequence capable of hybridising to the universal sequence of the oligonucleotide probe. A key element of this process is that it does not contain a normalisation step, but rather the initial step in generating the full length oligonucleotide for target amplification + adaptor addition uses a blocking group in the RC-PCR method, preventing the target-specific oligonucleotide from binding until the full length oligonucleotide is generated. While this can help with target uniformity, the lack of normalisation in this method can lead to some samples drowning out others and depleting a large amount of data during sequencing. This severely limits the number of targets that this method can accurately process. Referring to the Nimagen website (https: / / www.nimagen.com / applications / greenotype), this method (GREN OTYPE RC-PCR) is used for genotyping no more than 100 targets, while another method (GREN OTYPE MIP CAPTURE) is marketed for higher throughput (up to 5000 targets). Therefore, the RC-PCR method disclosed in GB2536446 is not suitable for processing a large number of targets.

[0016] Meldgaard et al. in Clinica Chimica Acta, Vol. 413; No. 19 (2012) and WO2017 / 044100 (Insilixa Inc.) both describe methods of target amplification and library preparation which use asymmetric PCR to generate single stranded products which are captured on printed arrays or beads, followed by analysis on a flow cytometer. Single stranded products are generated when one of the two primers in each set is in excess and the limiting primer is used up. The single stranded products are amplified in linear rather than exponential form. In both of these examples, the ultimate goal is to generate these single stranded products which can then be captured by hybridisation. Neither of these methods is suitable for targeting high throughput (NGS) sequencing or sample pooling.

[0017] Bybee et al. (Genome Biology and Evolution, Vol. 3, January 1, 2011) describe a method for generating a sequence library by a multiplex PCR method using single column barcodes with R®che 454 sequencing, which limits the number of samples that can be run at one time. The method requires a first PCR reaction in a first vessel before aliquoting the first reaction product into a second vessel for a second PCR step. The two-step PCR method is both laborious and time consuming, and transferring the product between vessels introduces the risk of cross contamination between samples. The method includes a bead immobilization step as part of the emulsion PCR step required for R®che 454 sequencing. In this method, each individual sample is quantified using pico-green, and then normalized using a liquid handling robot, which is both time consuming and expensive.

[0018] It is an object of the present invention to overcome at least one of the above problems. SUMMARY

[0019] The present invention provides a single closed tube method for preparing library constructs for targeted high throughput sequencing by multiplex amplification, without the need to re-open the tube to add additional components. The present invention also has built-in normalization, so that a separate normalization step is not required. With reference to Figure 1 , the method uses a target capture primer pair specific for each target in the sample, and a tagmentation primer pair specific for the sample. Each target capture primer pair comprises a forward and reverse primer comprising a target specific sequence and a read sequence, wherein the target specific sequence is configured to bind to the template nucleic acid at a position flanking the target (step 1). The tagmentation primer pair comprises a forward and reverse tagmentation primer, each primer having an adaptor sequence (P5, P7), an index sequence (xxxxx), and a sequence that binds to the read sequence (read sequence primer site) (step 2). The initial amplification step Figure 1 (1) produces an intermediate construct containing the target sequence flanked by read sequences (R1 and R2). In a subsequent amplification step Figure 1 (2), the tagmentation primers bind to the intermediate construct through the read sequence primer sequence that anneals to the read sequence. In the method of the present invention, one of the tagmentation primers is provided at a limiting concentration, and also includes a purification tag (B), which produces a reaction product containing partial and complete constructs Figure 1(3)). Partial constructs contain the target sequence, but only one adaptor sequence (P5) and one index sequence (xxxxx). On the other hand, full constructs contain the target sequence, first and second adaptor sequences (P5, P7) and first and second index sequences (xxxxx) and a purification tag. Because a limited amount of one tag primer (in the case shown, the reverse tag primer) is used, the reaction products contain an excess of partial constructs relative to full constructs, with the result that each individual reaction produces full sequencing constructs that have approximately the same amount of each targeted genetic locus (i.e., self-normalized) at relatively equal abundance across a wide range of template DNA. It will be appreciated that in Figure 1 the case shown, the read sequences R1, R2 and the adaptor sequences P5, P7 are for illustrative purposes only, and any read sequences, index sequences and adaptor sequences known to those skilled in the art can be used.

[0020] By using the method of the present invention, target capture, index and adaptor addition, followed by library capture using a purification tag can be performed in the same reaction vessel without opening the reaction vessel. After the amplification reaction, samples from thousands of individual reactions can be pooled together and the completed library constructs are captured and purified using the purification tag Figure 1 (biotin in the example). The unique combination of first and second index sequences (e.g., i5 and i7 barcode sequences) in each reaction allows for the pooling of all samples after multiplex amplification without the risk of cross-contamination. Only two unique tag primers are required per reaction, while the number of target-specific primers can vary depending on the desired assay panel. Initial amplification is performed using target capture primers that amplify the genomic DNA loci of interest. After the first two rounds of amplification, partial binding sites for the tag primers are generated, and they can begin to anneal to their complements. In the next few cycles, full-length complements of the tag primers are generated, and once this occurs, the remainder of the reaction is performed at a higher annealing temperature, reflecting the full priming site of the tag primers. The method of the present invention is suitable for multiplexing a plurality of samples (e.g., up to 200,000), wherein each sample contains thousands of target sequences (e.g., of SNPs), all of which can be performed in a single closed tube. For a sample with, for example, 50 target sequences, the method uses 50 target-specific primer pairs and one tag primer pair (specific to the sample). When the method is applied to a plurality of samples, for example, 500 samples, each with 50 target sequences, the method uses 50 x 500 target-specific primer pairs and 50 tag primer pairs (one for the sample). Amplification can be performed by polymerase chain reaction (PCR) or by other enzymatic nucleic acid amplification techniques.

[0021] In comparison to the method of Bybee et al., the method of the present invention allows the target capture primer pairs and the tagmentation primer pairs to be included in the same reaction vessel and incubated together simultaneously from the start and during the amplification step. This speeds up and simplifies the process and reduces the risk of cross-contamination, which is an inherent problem with 2-step PCR methods such as Bybee, where the reaction products of the first PCR step have to be cleaned up before aliquoting into the second reaction vessel for the second PCR step. Furthermore, the incorporation of the self-normalisation step in the method of the present invention eliminates the time-consuming and expensive requirement to quantify each sample individually and then normalise using a liquid handling robot.

[0022] In comparison to the method of GB 2536446, the incorporation of the self-normalisation step in the method of the present invention overcomes the problem of low throughput and allows the multiplexed amplification of hundreds or thousands of targets from multiple samples in a single run in a single vessel, regardless of the different samples having different concentrations or masses of target DNA, without compromising on target uniformity.

[0023] In a first aspect, the present invention provides a method of preparing library constructs for targeted next generation sequencing by multiplexed amplification, comprising the steps of:

[0024] (a) providing a first reaction vessel comprising:

[0025] (i) at least one sample, the sample comprising a plurality of different target sequences;

[0026] (ii) a plurality of target capture primer pairs for the at least one sample, wherein each target capture primer pair comprises:

[0027] a forward primer comprising, in the 5’ to 3’ direction, a first read sequence and a target-specific sequence; and

[0028] a reverse primer comprising, in the 5’ to 3’ direction, a second read sequence and a target-specific sequence;

[0029] (iii) a tagmentation primer pair for the at least one sample, comprising:

[0030] a forward tagmentation primer comprising, in the 5’ to 3’ direction, a first adaptor sequence, a first index sequence and a first read sequence primer site; and

[0031] a reverse tagmentation primer comprising, in the 5’ to 3’ direction, a second adaptor sequence, a second index sequence and a second read sequence primer site,

[0032] (b) performing sequential rounds of amplification at sequential annealing temperatures configured to amplify target sequences, yielding target sequences comprising first and second read sequences, and providing a reaction product comprising an adapter-ligated library of constructs in a sequential manner; and

[0033] (c) capturing adapter-ligated library constructs from the reaction product,

[0034] characterized in that one of the forward and reverse tagging primers comprises a purification tag at the 5' end and is provided at a limiting concentration, whereby the library of adapter-ligated constructs comprises:

[0035] partial constructs containing only one of the first and second index sequences and one of the first and second adapter sequences; and

[0036] full constructs containing the first and second index sequences, the first and second adapter sequences, and the purification tag,

[0037] wherein the reaction product comprises an excess of partial constructs relative to full constructs, and wherein step (c) comprises capturing only full constructs comprising the purification tag.

[0038] Typically, isolating (capturing) the full library constructs from the reaction product comprises contacting the reaction product with a support comprising a ligand for the purification tag.

[0039] The sample comprises or consists of nucleic acids containing target sequences. Typically, the sample is DNA. The target sequences are typically sequence variants, such as single nucleotide polymorphisms (SNPs) or short indels.

[0040] The capturing step typically comprises reacting the reaction product with a support comprising a ligand for the purification tag. The purification tag can be biotin, and the support can comprise streptavidin (e.g., streptavidin beads). Other purification tags and capture ligands can be used, the details of which are known to those skilled in the art.

[0041] The full library constructs can be released from the support for subsequent high-throughput sequencing, or it can be amplified in a multiplexed amplification step while still attached to the support to provide an amplification product, which can then be sequenced using high-throughput sequencing. In one embodiment, multiple amplification cycles are performed, and the amplification product is typically separated from the isolation support.

[0042] Typically, steps (a) and (b) are performed in a closed container (e.g., a tube), typically without opening the container prior to completion of these steps.

[0043] The thermal cycling of the amplification step is configured to amplify the target sequence in a sequential manner, to produce the target sequence comprising the first and second read sequences, and to provide a reaction product comprising the adapter-ligated library constructs in a sequential manner. In the described embodiment, using Illumina Rl and R2 read sequences, and i5 and i7 index sequences, the thermal cycling employs an initial, intermediate, and final number of amplification cycles, with increasing annealing temperatures. It will be appreciated that different thermal cycling methods can be used when using read sequences and index sequences from different sources. The use of Illumina read and index sequences is exemplary and is not intended to limit the scope of the application.

[0044] Thus, for example, the sequential number of amplification cycles at sequential annealing temperatures can comprise:

[0045] (i) performing one or more initial number of amplification cycles in the reaction vessel at a first annealing temperature;

[0046] (ii) performing one or more intermediate number of amplification cycles in the reaction vessel at a second annealing temperature configured to produce the target sequence comprising the first and second read sequences; and

[0047] (iii) performing one or more final number of amplification cycles in the reaction vessel at a third annealing temperature configured to provide a reaction product comprising the library of library constructs, the library of library constructs comprising partial constructs and complete constructs.

[0048] Typically, the second annealing temperature is higher than the first annealing temperature, and the third annealing temperature is higher than the second annealing temperature. However, different orders of annealing temperatures can be employed depending on the context.

[0049] In one embodiment, the first annealing temperature is 62°C ± 5°C, the second annealing temperature is 67°C ± 5°C, and the third annealing temperature is 72°C ± 5°C.

[0050] In one embodiment, the sequential number of amplification cycles comprises 1-5 initial number of amplification cycles (preferably 1-3 or 2 cycles), 1-5 intermediate number of amplification cycles (preferably 2-4 or 3 cycles), and 10-20 final number of amplification cycles (e.g., 10-50).

[0051] In one embodiment, the purification tag is biotin, and the separation step comprises reacting the reaction product with streptavidin beads.

[0052] In one embodiment, steps (a) and (b) are performed on a plurality of samples, wherein the reaction products of different samples are pooled, and capture / isolation step (c) is performed on the pooled reaction products. In one embodiment, steps (a) and (b) are performed on 2 to 200,000 samples, e.g. at least 100, 1000, 10,000, 20,000, 50,000, 100,000, 150,000 or 200,000 samples, and then pooled. In one embodiment, steps (a) and (b) are performed in the wells of a microtitre plate. In one embodiment, each sample comprises 1-10,000 target sequences, e.g. at least 10, 50, 100, 500, 1000, 2000, 3000, 4000, 5000, 6000, 7000, 8000, 9000 or 10,000 target sequences.

[0053] Amplification is typically performed by polymerase chain reaction (PCR). It will be appreciated that in order to perform PCR additional reagents are required and these will depend on the type of PCR employed. For example, in order to perform PCR the methods of the application use a thermostable DNA polymerase (e.g. Taq polymerase) and all four deoxyribonucleotides (dATP, dTTP, dCTP, dGTP).

[0054] In another aspect, the application provides a method of targeted high-throughput sequencing comprising the steps of:

[0055] The methods according to the application provide a library of complete constructs; and

[0056] The library of complete constructs is subjected to high-throughput sequencing.

[0057] In preferred embodiments, the high-throughput sequencing is a next generation sequencing technology. One example is sequencing by lllumination, in which the adaptor sequences (and typically the target primer pair and tag primer pair) are configured for sequencing by lllumination.

[0058] In one embodiment, the high-throughput sequencing has an on-target rate of at least 30%, 40%, 50% or 60%.

[0059] The methods of the application can be used to genotype samples for medical, diagnostic or commercial purposes. In one embodiment, the methods of the application can be used to genetically confirm the source or identity of a food product, e.g. a meat, fish, game, vegetable, legume, grain or fruit product.

[0060] In another aspect, the application provides a kit for preparing library constructs by multiplex amplification, suitable for targeted high-throughput sequencing of one or more samples, the kit comprising a tag primer pair for each sample, each tag primer pair comprising:

[0061] a forward tag primer comprising in the 5' to 3' direction a first adaptor sequence, a first index sequence, and a first read sequence primer site; and

[0062] a reverse tag primer comprising in the 5' to 3' direction a second adaptor sequence, a second index sequence, and a second read sequence primer site

[0063] wherein only one of the forward or reverse tag primers comprises a purification tag at the 5' end.

[0064] In one embodiment, the kit comprises:

[0065] a plurality of target-specific primer pairs (one per target sequence per sample), wherein each primer pair comprises:

[0066] a forward primer comprising in the 5' to 3' direction a first read sequence and a target-specific sequence; and

[0067] a reverse primer comprising in the 5' to 3' direction a second read sequence and a target-specific sequence;

[0068] In one embodiment, the purification tag is biotin.

[0069] In one embodiment, the kit comprises a support comprising streptavidin, such as streptavidin-coated beads or magnetic beads.

[0070] In one embodiment, the target primer pairs and tag primer pairs are configured for use in luminescent dye sequencing.

[0071] Other aspects and preferred embodiments of the application are defined and described in the other claims listed below. BRIEF DESCRIPTION OF DRAWINGS

[0072] Figure 1 . Closed tube multiplex library construction.

[0073] Stage 1 : Set up initial amplification comprising template DNA and target SNP sites marked with an asterisk. Forward and reverse target capture primers are marked for annealing to the template DNA at the positions flanking the target sequences by their respective Illumina target-specific sequences. Tag primers are also marked:

[0074] a forward tag primer comprising in the 5' to 3' direction a first adaptor sequence, a first index sequence, and a first read sequence primer site; and

[0075] a reverse tag primer comprising in the 5' to 3' direction a second adaptor sequence, a second index sequence, and a second read sequence primer site

[0076] Stage 2: After the initial low annealing temperature amplification cycle, partially formed DNA constructs hybridize to the pooled read sequences R1 and R2.

[0077] Stage 3: In the final stage, due to the limited amount of biotin-labeled reverse tag primer, amplification is allowed to proceed for many cycles, producing large amounts of partial constructs (containing the target sequence, R1 and R2 read sequences, one adapter sequence (P5) and one index sequence), while producing some amount of complete library constructs (containing the target sequence, R1 and R2 read sequences, two adapter sequences (P5 and P7) and two index sequences).

[0078] Figure 2 . Percent of reads between loci and forward primers containing the target sequence.

[0079] Figure 3 . Percent of reads between samples and summary data.

[0080] Figure 4 . Unit point scatter plot example. Each data point is one sample. DETAILED DESCRIPTION

[0081] All publications, patents, patent applications and other references mentioned herein are hereby incorporated by reference in their entirety for all purposes as if each individual publication, patent, patent application or other reference was specifically and individually indicated to be incorporated by reference for all purposes.

[0082] Definition and general preferences

[0083] As used herein, unless expressly stated otherwise, the following terms are intended to have the following meanings, as well as any broader (or narrower) meanings that the terms can enjoy in the art:

[0084] The singular shall be understood to include the plural, unless otherwise required by context. The term "one" shall be understood to refer to one or more of the referenced entity. Thus, the terms "one," "a," and "at least one" are used interchangeably herein.

[0085] As used herein, the term "comprises" or variations such as "comprising" will be understood to imply the inclusion of a stated integer (e.g., feature, element, property, characteristic, method / process step, or limitation) or group of integers (e.g., features, elements, properties, characteristics, method / process steps, or limitations) but not the exclusion of any other integer or group of integers. Thus, as used herein, the term "including" is inclusive or open-ended and does not exclude additional, unrecited integers or method / process steps.

[0086] A“target sequence” refers to a sequence, such as a variant of a nucleic acid sequence, such as a polymorphism (e.g., a single nucleotide polymorphism (SNP) or a short indel). The nucleic acid sequence can be any portion of a genomic nucleic acid, such as a coding or non-coding portion of a genome.

[0087] A“sample” refers to a composition comprising nucleic acids containing a target sequence. The sample is typically nucleic acid (e.g., DNA) purified from a biological material (e.g., tissue) sample. Other nucleic acids that can be amplified include cDNA. The DNA to be amplified can be obtained from any portion of any organism’s genome. For example, the DNA can be obtained from a human or other animal genome, a plant genome, a fungal genome, a bacterial genome, a viral genome, or any other DNA molecule. Methods for isolating nucleic acids from biological matter and tissue samples are well known to those of skill in the art. The methods of the present invention can be performed on a single sample comprising multiple targets, multiple samples comprising a single target, or multiple samples each comprising multiple targets. The methods of the present invention are particularly useful for use with multiple targets, such as greater than 100, 200, 300, 400, or 500 targets from one or more samples.

[0088] “High-throughput sequencing” refers to the sequencing of multiple different target sequences in parallel in the same reaction. The term includes“next-generation sequencing” or“NGS” (also known as“deep sequencing” or“massively parallel sequencing”), which is a DNA sequencing technology that enables millions of DNA small fragments to be sequenced in parallel. Various NGS technologies exist, including sequencing by synthesis (Illumina), pyrosequencing (454), ion semiconductor (Ion Torrent sequencing), and combinatorial probe anchor ligation (cPAS-BGI-MGI). Targeted NGS is a next-generation sequencing technology that focuses on amplicons and specific genes by amplifying the desired genes or amplicons through enzymatic amplification and then sequencing them on NGS platforms. It is described in Bybee et al. (“Targeted Amplicon Sequencing (TAS): A Scalable Next-Gen Approach to Multilocus, Multitaxa Phylogenetics”. Genome Biology and Evolution. 3: 1312-1323. doi:10.1093 / gbe / evr106. PMC 3236605) and Masser et al. (“Targeted DNA Methylation Analysis by Next-generation Sequencing”. Journal of Visualized Experiments. 96: 52488. doi:10.3791 / 52488. PMC 4354667).

[0089] A "library construct" refers to the product of the multiplex amplification reaction of the present application, which comprises a target sequence as well as first and second adaptor sequences to allow sequencing of the construct by next generation or high throughput sequencing (NGS) technology. The construct will typically also include first and second read sequences flanking the target sequence (e.g., Illumina Rl and R2 sequences), as well as two index sequences between the read sequences and the adaptor sequences (e.g., Illumina i5 and i7 sequences). The library will typically include partial constructs (which include only one adaptor sequence without the purification tag) as well as full (complete) constructs (which include the adaptor sequences, index sequences, and purification tag required for NGS). Capture technology using the purification tag allows for capture and use of the full constructs in NGS. Full (complete) and partial adaptor library constructs are illustrated in FIG. 3 (Stage 3). Figure 1

[0090] An "adaptor sequence" is a DNA sequence configured to allow sequencing of an adaptor-ligated library construct by next generation sequencing technology, such as any of the NGS technologies described herein, including Illumina dye sequencing in a flow cell. The adaptor sequence can be configured to allow sequencing of the adaptor-ligated construct by any NGS technology. Illumina's adaptors are described at https: / / support.illumina.com / downloads / illumina-adapter-sequences-document-1000000002694.html. Adaptor sequences for all major NGS platforms are provided by IDT (Integrated DNA Technologies) of Coralville, Ohio, USA (xGen™ UDI-UMI Adapters, xGen™ Stubby Adapters). IDT also provides a custom adaptor configurator tool that provides guidance for the design of custom NGS adaptors.

[0091] A "read sequence" is a portion of a target-specific primer. Both primers in a target-specific primer pair will include a read sequence. The read sequence is incorporated into the amplification product when the target DNA is amplified and the tagmentation primer can anneal to the sequence to which the tagmentation primer read sequence primer site. Examples of read sequences can include Illumina Rl and R2 read sequences. Read sequences are also available from Thermo Fisher as part of their Ion AmpliSeq TM DNA panels and library kits (https: / / www.thermofisher.com / ​

[0092] A “read sequence primer site” is a portion of a tagging primer that is complementary to one of the read sequences and allows the tagging primer to anneal to a target DNA comprising one of the read sequences.

[0093] The "index sequence" is part of the tagging primer that is introduced into the final library construct and acts as a barcode to identify the construct when analyzing sequencing data. Typically, the index sequence is 6-8 nucleotides in length and is located between the read sequence primer site and the adapter sequence in the tagging primer. The tagging primer pair will include two different index sequences, both of which will be incorporated into the complete (intact) library construct. An example of an index sequence is the Illumina i5 and i7 index sequences and the index oligonucleotides described in US2018334712. Suitable index sequences can be created using commercial products, such as:

[0094] https: / / rdrr.io / bioc / DNABarcodes /

[0095] https: / / omictools.com / dnabarcodes-tool

[0096] In the methods of the present invention, tagging primers containing a purification tag are provided at a "limited concentration". This means that the concentration provided results in insufficient tagging primers to bind to partially formed DNA constructs, resulting in an excess of partial constructs (lacking the purification tag) in the final product, while the complete (intact) construct containing the purification tag is limited. This, together with uniform initial target amplification, results in equal yields across samples, thereby avoiding the need for sample-specific downstream normalization. Figure 1 In the examples provided, reverse tagging primers containing the P7 adaptor sequence are provided in limited concentrations.

[0097] A "purification tag" refers to a tag that can be incorporated into the end of a primer sequence and can be used to purify the products of multiplex amplification. Typically, the tag is an affinity tag configured for purification of the amplified products using a support containing an affinity tag ligand (e.g., streptavidin beads). In one embodiment, the tag is biotin, the details of which are known to those skilled in the art. Other purification tags that can be used in the methods and products of the present invention include oligonucleotide tags (Acridite).

[0098] “On-target rate” refers to the percentage of returned sequence data specific to the target region. The higher the on-target rate, the more efficient the use of the sequencing run.

[0099] Example

[0100] The application will now be described with reference to specific examples. These are exemplary only and for illustrative purposes only: they are not intended to limit the scope of the patent or the described application in any way. These examples constitute the best mode presently contemplated for carrying out the application.

[0101] The present application provides a single tube method for targeted library preparation for next generation sequencing, which includes a normalization step using a purification tag such as biotin tag. Typically, multiplex library preparation comprises two steps: target capture, followed by tagging and adapter addition. In many methods of the prior art, these two steps are typically achieved by amplification. Typically, the two amplification steps are performed separately from the purification step. Hi-plex is a method that aims to combine these two amplification steps, but in Hi-plex, the index / adapter is introduced into the primer incorporation into the multiplex reaction after the initial amplification. This means that this method still requires opening and resealing of the sample container. This is not ideal for high-throughput processes (Nguyen-Dumont et al. 2013, Nguyen-Dumont et al. 2015, Pope et al. 2018, Hammet et al. 2019). The present application relates to a true single closed tube method, where target capture, indexing and adapter addition are performed in the same reaction vessel, followed by library capture using a purification tag, without the need to open and reseal the sample container, and without the need for a purification step prior to tagging and adapter addition.

[0102] The idea of the method of the present application is that each individual reaction will produce complete sequencing constructs, with the number of complete sequencing constructs per targeted locus being roughly the same, with equal relative abundance, across a wide range of input template DNA. The unique combination of index barcode sequences in each reaction allows for the pooling of all samples after multiplex amplification, without the risk of cross-contamination. The tagging primer, which is tagged using a low concentration of purification tag, allows the multiplex amplification to produce an excess of constructs containing only one adapter and index sequence, while limiting the amount of completed library constructs. This, together with uniform initial target amplification, enables the same output across samples, avoiding the need for sample-specific downstream normalization, and allowing multiplex amplification of hundreds to thousands of large number of targets. After the amplification reaction, samples from thousands of individual reactions can be pooled together, facilitating the capture and purification of complete library constructs through the purification tag, for example by using biotin as the purification tag and streptavidin beads to capture the biotin-tagged complete constructs. Figure 1 An overview of the single tube approach is shown. Unlike other methods, both the target capture primers and the tagmentation primers are contained in the same reaction vessel, and their order of activity is directed by the thermal cycling. While the number of target specific primers can vary depending on the panel desired, only two tagmentation primers are required per reaction. Initial amplification is performed using the target capture primers that amplify the genomic DNA sites of interest. After the first two rounds of amplification, partial binding sites for the tagmentation primers from the read sequences are generated, and they can begin to anneal to their complementary sequences. In the next few cycles, full-length complements of the tagmentation primers are generated, and once this occurs, the rest of the reaction proceeds at a higher annealing temperature that reflects the full binding sites of the tagmentation primers.

[0103] Current indicative cycling conditions are shown in Table 1.

[0104]

[0105] Table 1. Cycling conditions

[0106] Primer design

[0107] Target capture primers are designed such that the annealing temperature (Tm) of the target specific 3' end is between 59°C and 61°C. Primers are designed using a pipeline generated for multiplex amplification to avoid most sources of downstream primer heteroduplex artifact. Target capture primers carry the full-length read sequence, such as the Illumina Rl or R2 read sequence, matching the 3' end of the tagmentation primers. Each tagmentation primer contains a 6-base index barcode sequence sandwiched between an adaptor sequence, such as the Illumina p5 or p7 capture sequence, and a read sequence primer site. This allows for the use of dual-index sequencing, such as sequencing from Illumina, and allows for demultiplexing of sequences from genetic data, such as fastq data, for downstream genotyping of individual samples. For example, the p5 end of a construct, the tagmentation primer can consist of the p5 sequence, a unique 6-base barcode sequence, and the Illumina read 1 sequencing primer site. While for the p7 end of a construct, the tagmentation primer consists of a biotin-labeled 5' end, the p7 sequence, a unique 6-base barcode sequence, and the Illumina read 2 sequencing primer site. The overlapping portions of the primers typically have an annealing temperature of about 60°C. Once the full-length complements are generated, the Tm for the complete primer site for each end of a construct is about 74°C.

[0108] Normalization

[0109] The method does not require an isolating normalization step after library construction. Instead, the same representation of different samples is achieved by controlling the concentration of the tag-labeled primers in the cycling conditions and the limiting purification. The tag-labeled primers are depleted at the later stages of the thermal cycling, allowing an equal amount of library complete constructs to bind to the support through the added purification tag; the total amount of construction can vary between samples, but the fraction containing the complete tag should be similar. Therefore, all samples can be pooled together after library construction and all downstream steps are performed on a single sample; library capture using a purification support, library release from the support (e.g. with short thermal cycling), final library cleanup (e.g. using magnetic beads) and library quantification. A suitable tag-labeled primer is the i7 tag-labeled primer from Illumina. A suitable purification tag is the biotin tag, which can be captured with streptavidin beads.

[0110] Materials

[0111] The DNA used in the experiment was extracted from a pig ear tag tissue sample using magnetic beads, with an average yield of 24 ng / μl (SD 11 ng / μl). All target capture primers were ordered from IDT, synthesized at a scale of 25 nmoles, and used at a concentration of 200 uM in Tris-EDTA pH 8.0 buffer with standard desalting purification. The biotin-labeled i7 tag-labeled primer was lyophilized in a test tube and resuspended in nuclease-free water to a stock concentration of 10 uM.

[0112] Multiplex amplification

[0113] Target capture primers were pooled and diluted to a concentration of 0.5uM per primer as a working stock. For the test library, 51 target capture primers were used to create the primer pool. Individual reactions were set up in a total volume of 7uL (3.5uL Qiagen Plus Multiplex Mastermix, 2uL DNA extract, 0.5uL target capture primer mix, 1uL 10uM i5 tag primer, ~0.01uL biotinylated i7 tag primer). The amount of i5 tag primer required varied depending on the number of target capture primers present in the pool, so that the molar amount of available primer sites was approximately equal to the molar amount of i5 primer. In this case, the final concentration of all i5 tag primer sites was ~1.8uM in the final reaction conditions, and the concentration of added i5 tag primer used was ~1.4uM. Amplification was set up in a 96 well plate, starting by mixing 52uL of pooled target primer mix, 1uL of 10uM biotinylated i7 tag primer and 371uL of Qiagen plus multiplex mastermix (Qiagen, UK) to make the master mix. 4uL of master mix was dispensed into each well, followed by 1uL of 10uM i5 tag primer and 2uL of template DNA using a multichannel pipette. The plate was heat sealed, vortexed gently and centrifuged briefly. Amplification was carried out under the following cycling conditions (95°C - 15m (1x hot start); 95°C - 30s, 61°C - 30s (slow ramp 0.2°C / s), 72°C - 30s (5x); 95°C - 30s, 67°C - 30s (x3); 95°C - 30s, 72°C - 30s (x22); 4°C - hold).

[0114] Library capture and normalisation

[0115] Following amplification, 4uL of each reaction was pooled into one pool using a multichannel pipette. A 500uL aliquot was combined with 500uL of 2x binding buffer (10mM Tris-Hcl (pH 7.5), 1mM EDTA, 200mM NaCl and 0.02% Tween20 buffer). One microlitre of streptavidin beads was washed in 1mL of 1x binding buffer in a 1.5mL tube and captured using a magnetic stand. The supernatant was discarded, the tube removed from the magnetic stand and the magnetic beads resuspended using 1mL of the pooled amplification mixture. The beads were incubated at room temperature for 15 minutes, then returned to the magnetic stand for approximately 3 minutes until the supernatant was clear. The supernatant was removed on the magnetic stand and the beads washed once using 1mL of 1x binding buffer. The supernatant was again discarded and the beads resuspended in 20uL of elution buffer (10mM Tris-HCl pH 8.0).

[0116] To release the captured library constructs from the beads, another amplification was performed using the resuspended streptavidin beads. A single tube reaction was set up by mixing 20uL Qiagen plus master mix (any PCR master mix can be used), 4uL of 10uM Illumina P5 primer, 4uL of 10uM Illumina P7 primer and 12uL of resuspended streptavidin beads. Amplification was performed using the following thermal cycling conditions (94°C - 15m (hot start); 94°C - 30s, 60°C - 30s, 72°C - 30s (5x); 4°C - hold). After 5 amplification cycles, the mixture was transferred to a new 1.5mL tube and placed on a magnetic stand. 25uL of the clear supernatant was transferred to a new 1.5mL tube and mixed with 15uL CleanNGS (CleanNA, Netherlands) beads. The mixture was incubated at room temperature for 5 minutes before the tube was returned to the magnetic stand for 3 minutes. The clear supernatant was then transferred to a new 1.5mL tube and mixed with a further 13uL CleanNGS beads. The mixture was again incubated at room temperature for 5 minutes before being placed on the magnetic stand for 3 minutes. The supernatant was removed and the beads were washed twice using 200uL fresh 70% ethanol while still on the magnetic stand. The ethanol was removed and the tubes were air dried for 10 minutes to allow the residual ethanol to evaporate. The beads were then resuspended in 15uL elution buffer and the beads were captured on a magnetic stand. The clear supernatant was collected and transferred to a new 1.5mL tube and 1.5uL of elution buffer containing 1% Tween 20 was added. This was the final defined library.

[0117] Library quantification

[0118] The concentration of the final library was determined using the Illumina library quantification kit from Kapa Biosystems (Roche Sequencing, USA) according to the manufacturer’s instructions for an Applied Biosystems StepOnePlus instrument (thermo fisher scientific, UK).

[0119] Sequencing and data analysis

[0120] Sequencing of completed libraries was performed using an Illumina MiSeq instrument using the paired-end 75 cycle reagent kit. Following sequencing, data was converted to fastq format using Illumina conversion scripts that allow for the addition of i7 and i5 index sequences to the header line of each sequence. Sequencing reads were split into individual fastq files for each individual sample using a python script as GTseq_BarcodeSplit_MP.py that provides the expected barcode combinations as input files. Each individual fastq file was then used for genotyping using a perl script as GTseq_Genotyper_v3.pl. Summary files containing genotypes, allele ratios, and other sequence read data were then analyzed to assess the efficiency of the multiplexed amplification library preparation method.

[0121] Results

[0122] The optimal performance conditions were tested on a set of 24 samples. Four samples were removed from analysis because these samples contained no template, reagents evaporated during PCR, or performed poorly under all tested conditions. For the subset of samples representing the proposed conditions, 835,017 raw reads containing the expected 6 base barcode combinations were returned following sequencing. Evidence of evaporation wells due to poor heat sealing was noted prior to sequencing, and as expected, these wells returned poor numbers of reads. Since these samples did not represent typical reaction conditions, these samples were removed from further analysis. Raw reads from each of the remaining individual samples (n=20) were relatively uniform, with an average of 41,751 reads and a standard deviation of 18,068. The number of raw reads returned ranged from 26,890 to 93,038. The percentage of targeted sequences (as opposed to non-targeted sequences or artificial sequences - also referred to as "on-target ratio") averaged 31% across the 20 analyzed samples. Even with the lower number of targeted reads, the reads from all 20 analyzed samples were sufficient to well genotype, with an average call rate of 98.9% and all call rates above 98%. The uniformity of reads across loci meant that a low number of reads were needed to reach high call rates, and only 2,400 on-target reads were needed to meet a 90% threshold Figure 3 ). The selected target loci performed as expected and produced clean allele ratios that were easily scored using the genotyping pipeline Figure 4 ).

[0123] Equivalent plan

[0124] The foregoing description details a current preferred embodiment of the application. It will be appreciated that those skilled in the art will be able to devise many modifications and variations without departing from the application. Such modifications and variations are intended to be included within the scope of the application as defined by the appended claims. For the sake of brevity, the embodiments are described as single embodiments, however, it will be appreciated that various combinations of these embodiments are within the scope of the application.

[0125] References

[0126] Hammet, F., Mahmood, K., Green, T. R., Nguyen-Dumont, T., Southey, M. C, Buchanan, D. D., Lonie, A., Nathanson, K. L., Couch, F. J., Pope, B. J. and Park, D. J. 2019: Hi-Plex 2: a simple and robust approach to targeted sequencing-based genetic screening. BioTechniques. 67(3): 00-00 (September 2019). 10.2144 / btn-2019-0026.

[0127] Nguyen-Dumont, Tu, Pope, B., Hammet, F., Southey, M. and Park, D. 2013: A high-plex PCR approach for massively parallel sequencing. Biotechniques. 55: 69-74.

[0128] Nguyen-Dumont, Tu, Pope, B., Hammet, F., Southey, M. and Park, D. 2013: A high-plex PCR approach for massively parallel sequencing. Biotechniques. 55: 69-74.

[0129] Pope, B., Hammet, F., Nguyen-Dumont, T. and Park, D. 2018: Hi-Plex for simple, accurate, and cost-effective amplicon-based targeted DNA sequencing. Chapter 5 in Steven R. Head et al. (eds), Next Generation Sequencing: Methods and Protocols. vol. 1712.

Claims

1. A method for preparing a library construct for targeted next generation sequencing by multiplex amplification, comprising the following steps: (a) providing a first reaction vessel comprising: (i) at least one sample, wherein the sample comprises a plurality of different target sequences; (ii) a plurality of target-specific primer pairs for the at least one sample, wherein each primer pair comprises: a forward primer comprising the first read sequence and the target-specific sequence in the 5' to 3' direction; and a reverse primer comprising the second read sequence and the target-specific sequence in the 5' to 3' direction; (iii) a tagging primer pair for the at least one sample, comprising: a forward tagging primer comprising a first adaptor sequence in the 5' to 3' direction, first index sequence and first read sequence primer sites; and a reverse tagging primer comprising a second adaptor sequence in the 5' to 3' direction, a second index sequence and a second read sequence primer site, (b) performing sequential rounds of amplification in a first reaction vessel at a sequential annealing temperature, the temperature being configured to amplify the target sequence, generate the target sequence comprising the first and second read sequences, and provide a reaction product comprising a library of adaptor-ligated constructs in a sequential manner; and (c) capturing a library of adaptor-ligated constructs from said reaction products, Characterized in that one of the forward and reverse tagging primers comprises a purification tag at the 5' end and is provided in a limited concentration, whereby the library of adaptor-ligated constructs comprises: a partial construct containing only one of the first and second index sequences and one of the first and second adapter sequences; and a complete construct comprising the first and second index sequences, the first and second adaptor sequences, and a purification tag, wherein the reaction product comprises an excess of the partial construct relative to the complete construct, and wherein step (c) comprises capturing only the complete construct comprising the purification tag.

2. The process of claim 1, wherein the first reaction vessel is closed during step (b).

3. The method of claim 1 or 2, wherein the sequential rounds of amplification at sequential annealing temperatures comprise: (i) performing one or more initial rounds of amplification in the reaction vessel at a first annealing temperature; (ii) performing one or more intermediate rounds of amplification in the reaction vessel at a second annealing temperature, wherein the second annealing temperature is configured to generate a target sequence comprising the first or second read sequence; as well as (iii) performing one or more final rounds of amplification in the reaction vessel at a third annealing temperature, wherein the third annealing temperature is configured to provide a reaction product comprising a uniform amount of the complete library construct. The method of claim 3 , wherein the second annealing temperature is higher than the first annealing temperature, and the third annealing temperature is higher than the second annealing temperature. 5 . The method of claim 4 , wherein the first annealing temperature is 61° C.±5° C., the second annealing temperature is 67° C.±5° C., and the third annealing temperature is 72° C.±5° C.

6. The method according to any one of claims 3 to 5, comprising 1-5 initial rounds of amplification, 1-5 intermediate rounds of amplification and 10-20 final rounds of amplification.

7. The method according to any one of claims 1 to 6, wherein the purification tag is biotin and the capturing step comprises reacting the reaction product with streptavidin beads.

8. A method according to any of the preceding claims, wherein steps (a) and (b) are performed on a first sample in the first reaction vessel to produce a first reaction product comprising a first library of adaptor-ligated constructs, and steps (a) and (b) are performed on a second sample in a second reaction vessel to produce a second reaction product comprising a second library of adaptor-ligated constructs, wherein the first and second reaction products are combined and the capture step (c) is performed on the combined reaction products.

9. The method of claim 8, wherein steps (a) and (b) are performed on each of more than 100 samples.

10. The method of claim 8, wherein steps (a) and (b) are performed on each of more than 1000 samples.

11. The method according to any preceding claim, wherein each sample comprises at least 10 target sequences.

12. The method according to any preceding claim, wherein each sample comprises at least 50 target sequences.

13. The method according to any one of the preceding claims, wherein the capturing step comprises capturing the complete construct on a support and subsequently amplifying the complete construct attached to the support.

14. A method for targeted next generation sequencing, comprising the following steps: Providing a library of complete constructs according to the method of any preceding claim; as well as A library of complete constructs was subjected to high-throughput sequencing.

15. The method of claim 14, wherein the next generation sequencing is luminescent dye sequencing.

16. A kit for preparing a library construct by multiplex amplification, wherein the library construct is suitable for targeted high-throughput sequencing of at least one sample containing multiple target sequences, the kit comprising: (ii) a plurality of target-specific primer pairs for each sample, wherein each primer pair comprises: a forward primer comprising the first read sequence and the target-specific sequence in the 5' to 3' direction; and a reverse primer comprising the second read sequence and the target-specific sequence in the 5' to 3' direction; as well as (iii) a tagging primer pair for each sample comprising a forward tagging primer comprising, in the 5' to 3' direction, a first adapter sequence, a first index sequence, and a first read sequence primer site; as well as a reverse tagging primer comprising, in the 5' to 3' direction, a second adapter sequence, a second index sequence, and a second read sequence primer site, Only one of the forward or reverse tagging primers contains a purification tag at the 5' end.

17. The kit of claim 16, wherein the purification tag is biotin.

18. The kit according to claim 16 or 17, comprising streptavidin magnetic beads.

19. The kit of any one of claims 16 to 18, wherein the target primer pair and tagging primer pair are configured for use with luminescent dye sequencing. 20 . The kit according to claim 16 , which is used for multiplex amplification of a plurality of samples and comprises a plurality of target-specific primer pairs for each sample, and a tagging primer pair for each sample.

21. A library of adaptor-ligated constructs obtainable by the method according to any one of claims 1-15.

Images