JP2013165661A - Method for determining base sequence of two or more specimens at a time by corresponding each specimen to sequence - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method for determining base sequences of two or more specimens at a time by corresponding each specimen to sequences.SOLUTION: Sequences are determined at a time through correspondence to each specimen by preparing a plurality of groups by mixing two or more nucleic acid specimens in a specific combination, adding an index sequence specific to each group, subsequently carrying out a sequential analysis using the next generation sequencer (or massively parallel sequencer, NGS), corresponding to each specimen based on the combined pattern of the added index sequence in sequences whose specimen-corresponding parts are the same, and assembling the sequence groups already corresponded to the same specimen.

Description

The present invention creates a subgroup in which a plurality of nucleic acid samples are mixed in a specific combination, assigns an index sequence unique to each subgroup, performs sequence analysis using a massively parallel sequencer, and converts the obtained sequence into an index. The present invention relates to a method of associating and classifying each sample and a sequence by assembling those classified according to the sequence.

Although the next-generation sequencer can output a large amount of sequences at a time by parallel processing, the sample and the output sequence cannot be associated with each other because the sample is mixed and reacted. Therefore, it is used for analysis aiming at clarifying the population such as analysis of DNA and RNA derived from one specimen such as genome analysis and transcriptome analysis, and metagenome analysis (FIG. 1).

On the other hand, in order to effectively use the analysis power of the next-generation sequencer, when processing a sample, an artificial base sequence (Index) consisting of several bases to a dozen bases is assigned to associate the sample with the output sequence, A multiplex method is known as a method for performing multi-sample processing, but the number of parallel samples is limited to several tens of samples.

Cell Engineering August 2011 Issue Mastering the Next Generation Sequencer Basics Cell Engineering August 2011 Issue Using the Next Generation Sequencer De novo DNA sequencing of bacterial genome by multiplex method Cell Engineering August 2011 Issue Using the Next Generation Sequencer De novo DNA sequencing of bacterial genome by multiplex method Parasitol Int. 2011 Jun; 60 (2): 199-202. Epub 2011 Mar 21. Construction and analysis of full-length cDNA library of Cryptosporidium parvum. Genome Biol. 2009; 10 (3): R25. Epub 2009 Mar 4. Ultrafast and memory-efficiency alignment of short DNA sequences to the human genome.

An object of the present invention is to provide a method of achieving multiple samples exceeding the multiplex method by improving the Index providing method and effectively utilizing the analysis power of the next-generation sequencer.

As a result of earnest examination of the index assignment method and the analysis method of the sequence output from the next-generation sequencer, the present inventor mixed each sample in a specific combination and then compressed it to a number proportional to the root of the number of original samples. An index is assigned to each mixed sample, a base sequence is obtained by a next-generation sequencer, and a method of restoring the sequence before mixing in association with the ID of the specimen from the output sequence has been found, and the present invention has been completed (FIG. 2). ).

That is, this invention consists of the following aspects.

(1) A plurality of IDs (1, 2, 3, 4) are assigned to each of a plurality of nucleic acid samples (plasmids A, B, C, D) so as not to have the same combination (A = 1 & 2) , B = 1 & 3, C = 2 & 4, D = 3 & 4), and a sub-group is formed by combining samples having the same ID (division / mixing: 1 = A + B, 2 = A + C, 3 = B + D , 4 = C + D), each subgroup is fragmented, and a subgroup-specific index sequence (1 = AA, 2 = CT, 3 = CG, 4 = AT, Index binding) is added to the nucleic acid constituting each subgroup. After adding, all are mixed and sequence analysis is performed (mixed sequence), and the obtained sequence is classified according to the Index sequence to associate with each specimen (grouping according to Index: A = AA + CT, B = AA + CG, C = AT + CT, D = AT + CG), determine the base sequence of each sample by assembling the associated sequences How. (The figure in parentheses shows the case of FIG. 2 illustrating the simultaneous analysis of 4 samples)

(2) A method for determining a base sequence, wherein the number of specimens in (1) is n specimens of a composed of n0 specimens and (a article n-n0) mock specimens.

(3) The subgroup configuration method in (2) is the same for each digit, assuming that (X-1) is expressed in a-adic for the Xth specimen, and the combination of each digit and each numeric value is regarded as an ID. A method for determining a base sequence, comprising an operation that is synonymous with repeating for all the digits that the samples having the numerical values are collectively made into one subgroup.

(4) A method for determining a base sequence, wherein a in (3) is 2 or a multiple of 2.

(5) The method of associating the obtained sequence in (4) with each sample by classifying according to the Index sequence is a combination of various Index sequences given to the same sequence and given to the sample A method for determining a base sequence, which is performed by collating with a combination of assigned IDs.

(6) A method for determining a base sequence, wherein the nucleic acid sample in (1) is DNA or RNA.

(7) A method for determining a base sequence, wherein the sequence analysis technique in (1) above uses a next-generation sequencer.

(8) A method for determining a base sequence, wherein the Index sequence in (1) is DNA, RNA or a mixed substance thereof.

(9) A method for determining a base sequence, wherein the Index sequence in (8) includes a parity factor.

(10) A method for determining a base sequence, wherein the methods described in (6) to (9) are arbitrarily combined with the above (1) to (5).

(11) A computer program for executing the method according to (11) above by a computer.

According to the present invention, it is possible to determine the sequence of the number of samples corresponding to the power of the number of indexes to be used by one-time sequencer analysis (FIGS. 1 and 2).

For example, by using 20 types of Index sequences, the array of 2 to the 10th power, that is, 1,024 samples, and by using 40 types of Index sequences, the sequence of 2 to the 20th power, that is, the sequence of 1,048, 576 samples. Can be determined by one-time next-generation sequencer analysis while maintaining correspondence information between the specimen and the output sequence.

Specifically, it is possible to replace the base sequence analysis of plasmid DNA, which has been performed by the capillary method so far, with the combination of the present invention and the next-generation sequencer, and as a result, the cost-effectiveness can be greatly improved. It is.

Effect of the invention Summary of invention Overview of sample mixing method What is parity Summary of Invention (2) Accession number of the base sequence used in Example (1) Composition of mixed group used in Example (1) Example (1) Restoration rate Example (1) Restoration rate

Hereinafter, embodiments of the present invention will be described in detail.

In the present invention, the next-generation sequencer is a group of base sequence analyzers that have been developed in recent years, and whose analysis capabilities have been dramatically improved by parallel processing. Although it refers to Ion PGM, Pacific Biosciences, etc. (Non-patent Document 1), it includes devices to be developed in the future and is not limited to those shown here.

In the present invention, a nucleic acid sample is an assembly of DNA or RNA molecules having the same sequence. That is, a plasmid amplified and cloned in a system using E. coli is applicable. The nucleic acid sample may be a PCR product. The above-mentioned identical sequence includes not only a completely matched sequence but also a collection of molecules including partially identical sequences. Furthermore, a certain amount of contamination may be included.

The method for mixing nucleic acid samples is illustrated in FIG. 3 and the following, but is not limited thereto.

When the number of specimens is n0, a is an integer constant, and the minimum number n of a to the power of n> n0 is obtained. By adding (a to the power of n−n0) mock specimens, N-th sample of the n-th power of 1) is prepared, (X-1) is expressed in a-adic number for the X-th sample, and a combination of each digit and each numerical value is regarded as an ID, Samples having the same numerical value for each digit are mixed together as one subgroup.

Although the specific example in the case of a = 2 is shown with Table 1 below, the mixing method is not limited to the illustrated method.

When nucleic acid samples consisting of 2 n power types from # 0 to # 2 n-1 are analyzed in parallel, specimens are mixed based on the following rules.

Specimen number is expressed in binary number. If the numerical value of the k-th digit is 0, a mixed group ID (2 × (n−k) +1) is given, and if it is 1, a mixed group ID (2 × (n−k) +2) is given. This is repeated for all k satisfying 1 to n.

For example, when n = 3, the total number of samples is 8, the sample number # 6 is expressed as 110 in binary, and the third digit is 1, so (2 × (3-3) +2 ) = 2, since the number of the second digit is 1, (2 × (3-2) +2) = 4, since the number of the first digit is 0, (2 × (3-1) +1) = 5, and the result The sample numbers # 6 are added with the mixed group IDs 2, 4, and 5. On the other hand, the samples to which ID2 is added are sample # 6 (110), sample # 7 (111), sample # 5 (101), and sample # 4 (100), so a mixed sample in which these are collected is created.

A nucleic acid sample composed of 2n types is organized into 2n types of mixed samples according to the above rules, and then subjected to a known multiplex method using a next-generation sequencer (Non-patent Document 2) to obtain a base sequence.

A parity base can be added to the Index used at this time (FIG. 4).

The base sequence is reconstructed as follows (FIGS. 2 and 5).

An index is assigned to the sequence output from the next-generation sequencer, and the sequence and the mixed ID can be associated with each other based on the index.

When parallel analysis is performed after 2n types of nucleic acid samples are organized into 2n types of mixed samples, for example, sample numbers # 0 are included in the mixed IDs 0, 2, 4,..., 2n-2. Similarly, the sample numbers # 1 are included in the mixed IDs 0, 2, 4,..., 2n-1.

That is, the index corresponding to the mixed IDs 0, 2, 4,..., 2n-2 is assigned to the sequence derived from the specimen number # 0. Similarly, an index corresponding to the mixed IDs 0, 2, 4,..., 2n-1 is assigned to the sequence derived from the specimen number # 1.

On the contrary, the original specimen number can be specified by counting the Indexes assigned to the same sequence and analyzing the distribution.

For example, when an index corresponding to the mixed ID 1, 3, 5,..., 2n-2 is assigned to a certain base sequence output from the next-generation sequencer, the sequence is a sequence constituting the sample number # 0. I understand that there is. Similarly, when an index corresponding to the mixed IDs 2, 4, 5,..., 2n-1 is assigned to a certain sequence, it is understood that the sequence is a sequence constituting the specimen number # 1.

The base sequence of each sample can be obtained by creating a contig with an assembler based on the output sequence from the next-generation sequencer associated with the original specimen number. Here, the assembler may use ABySS (Non-Patent Document 3) or Velvet, but is not limited thereto.

The output sequence from the next-generation sequencer may be appropriately shortened and used for the above analysis.

Hereinafter, the embodiments of the present invention will be described with reference to examples. However, this is merely an example and does not limit the present invention.

(Example 1) Simulated analysis by computer A known full-length cDNA sequence of Cryptosporidium parvum consisting of 1,066 sequences was used as a target sequence for simulation analysis. This sequence includes sequence fragments revealed by the Sanger method and 18,308,250 sequence fragments (accession numbers: SRX004536 to SRX004538) revealed by shotgun analysis using a next-generation sequencer. It was determined by comparing with the genome sequence of known Cryptosporidium parvum (Non-patent Document 4).

In this example, a in claim 2 was set to 2, n was set to 8, and 256 sequences were selected from the 1,066 sequences (FIG. 6). In addition, a 5,474,164 fragment sequence corresponding to 256 sequences was selected from the 18,308,250 fragment sequences and used thereafter.

In this example, it was attempted by simulation analysis using a computer to reconstruct 256 sequences from the 5,474,164 fragment sequences by the processing based on the present invention.

Sixteen groups were constructed by mixing the 256 sequences of interest in the combinations shown in FIG. That is, each array belongs to 8 types of groups.

By using Bowtie (Non-Patent Document 5) which is a mapping tool, the 5,474,164 fragment sequences were associated with 256 sequences of interest, and these were randomly assigned to the 8 groups. This is an operation peculiar to the simulation analysis, and is replaced with a process based on the Index array in the actual analysis. In addition, 4 bases on the 5 'side were removed as an Index-corresponding sequence, and the remaining 32 bases were used for the subsequent analysis.

Subsets of 11 sequences were created from 1 sequence by extracting the 32 bases to be analyzed in units of 22 consecutive bases.

By counting 22 base sequences having the same sequence and analyzing the index sequence pattern assigned to them, it was estimated which sequence of the 256 bases was derived from the 22 bases.

For each of the 256 sequences, the 22-base sequence corresponding to each of the 256 sequences was tabulated, and a contig was created by the assembler ABySS (Non-patent Document 3).

The degree of restoration was estimated by comparing the generated contig sequence with the original sequence.

Here, the degree of restoration is defined by (length of maximum length contig / length of original array).

As a result, it was shown that most of the sequences can be restored by this method. That is, of the 256 sequences, 104 sequences can restore 99% or more of the region, 160 sequences can restore 95% or more of the region, and 199 sequences can restore 90% or more of the region. A grade was obtained. Details are shown in FIGS.

It is possible to replace the base sequence analysis of plasmid DNA, which has been performed by the capillary method so far, with a combination of the present invention and a next-generation sequencer, and as a result, a significant improvement in cost effectiveness can be expected.

Claims (11)

A plurality of IDs (1, 2, 3, 4) are assigned to each of a plurality of nucleic acid samples (plasmids A, B, C, D) so as not to have the same combination (A = 1 & 2, B = 1 & 3, C = 2 & 4, D = 3 & 4), and a sub-group is formed by combining samples having the same ID (division / mixing: 1 = A + B, 2 = A + C, 3 = B + D, 4 = C + D), each subgroup is fragmented, and subsequence-specific index sequences (1 = AA, 2 = CT, 3 = CG, 4 = AT, index binding) are added to the nucleic acids constituting each subgroup. All are mixed and sequence analysis is performed (mixed sequence), and the obtained sequences are classified according to the index sequence to associate with each specimen (grouping according to the index: A = AA + CT , B = AA + CG, C = AT + CT, D = AT + CG), determining the base sequence of each specimen by assembling the associated sequences . (The figure in parentheses shows the case of FIG. 2 illustrating the simultaneous analysis of 4 samples) The base sequence determination method according to claim 1, wherein the number of samples is n0 samples and (n-th power of n-n0) mock samples. The subgroup configuration method according to claim 2 has (X-1) expressed in a-adic for the X-th specimen, and a combination of each digit and each numerical value is regarded as an ID, and each digit has the same numerical value. A method for determining a base sequence, comprising an operation synonymous with repeating all the digits to group samples into one subgroup. The method for determining a base sequence, wherein a in claim 3 is 2 or a multiple of 2. The method of associating each sequence obtained by classifying the obtained sequence according to claim 1 with each sample is a combination of various index sequences assigned to the same sequence and the ID assigned to the sample. A method for determining a base sequence, which is performed by matching with a combination. The method for determining a base sequence, wherein the nucleic acid sample according to claim 1 is DNA or RNA. A method for determining a base sequence, wherein the sequence analysis technique according to claim 1 uses a next-generation sequencer. The method for determining a base sequence, wherein the Index sequence according to claim 1 is DNA, RNA or a mixed substance thereof. The method for determining a base sequence, wherein the Index sequence according to claim 8 includes a parity factor. A method for determining a base sequence, wherein the methods according to claims 6 to 9 are arbitrarily combined with claims 1 to 5. A computer program for executing the method according to claim 10 by a computer.