Neoantigen prediction device, neoantigen prediction method, and program

Abstract

To solve a problem that a neoantigen could not be highly accurately predicted, and positional information of a neoantigen could not be highly accurately acquired.SOLUTION: By a neoantigen prediction device A comprising: a subject reception part 21 which receives one or more reads of subject genome information; a mapping part 31 which acquires the positional information of one or more each read from reference genome sequence information; a determination part 33 which acquires 2 or more kmer from one or more each read, and determines one or more kmer which is not present in normal DB out of the 2 or more kmer; a position acquisition part 35 which acquires the positional information of amino acid sequence information corresponding to one or more each kmer; a neoantigen acquisition part 36 which acquires neoantigen information being information in which the positional information acquired by the position acquisition part 35 is associated with one or more each amino acid sequence information; and a neoantigen output part 41 which outputs the neoantigen information, a neoantigen can be highly accurately predicted, and the positional information of a neoantigen can be highly accurately acquired.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a neoantigen prediction device that outputs information regarding neoantigens. [Background technology]

[0002] In the prior art, there are techniques for prioritizing candidate neoantigens for a patient (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2018-524008 Summary of the Invention [Problem to be solved by the invention]

[0004] However, conventional techniques have been unable to predict neoantigens with high accuracy, and have been unable to obtain position information of neoantigens with high accuracy. [Means for solving the problem]

[0005] The neoantigen prediction device of the first invention is a neoantigen prediction device comprising: a target receiving unit that receives one or more reads that are a first length portion of target genome information including abnormal genome information; a mapping unit that acquires position information that is information that identifies the position of each of the one or more reads by referring to reference genome sequence information that is normal genome information and from which position information of the reads that constitute the genome information is acquired; a determination unit that acquires two or more kmers that are a second length (first length >= second length) portion from each of the one or more reads, and determines one or more kmers among the two or more kmers that do not exist in a normal database in which the two or more kmers acquired from the normal genome information are stored; a position acquisition unit that acquires position information of amino acid sequence information corresponding to each of the one or more kmers determined by the determination unit, using the position information acquired by the mapping unit, which is position information of the reads corresponding to the kmers corresponding to the amino acid sequence information; a neoantigen acquisition unit that acquires neoantigen information that is information that associates the position information acquired by the position acquisition unit with each of the one or more amino acid sequence information; and a neoantigen output unit that outputs the neoantigen information.

[0006] With this configuration, neoantigens can be predicted with high accuracy, and position information of neoantigens can be obtained with high accuracy.

[0007] In addition, compared to the first invention, the neoantigen prediction device of the second invention further includes a range receiving unit that receives range information that specifies the range to be examined in the target genome information, and a selection unit that uses position information of the reads acquired by the mapping unit to select one or more reads that correspond to the position of the range specified by the range information, and the determination unit is a neoantigen prediction device that determines one or more kmers that do not exist in the normal database from the one or more reads selected by the selection unit.

[0008] With this configuration, neoantigens can be predicted more accurately and quickly, and position information of neoantigens can be obtained with high accuracy.

[0009] In addition, the neoantigen prediction device of the third invention is a neoantigen prediction device in which, compared to the first or second invention, the determination unit comprises a kmer acquisition means for acquiring one or more kmers from each of one or more reads, a determination means for determining one or more kmers from the one or more kmers acquired by the kmer acquisition means that do not exist in the normal database, a counting means for acquiring the number of occurrences in each of the one or more reads for each of the one or more kmers, and a kmer selection means for selecting one or more kmers whose number of occurrences satisfies a number of occurrences condition from the one or more kmers determined by the determination means.

[0010] With this configuration, neoantigens can be predicted quickly and accurately, and position information of neoantigens can be obtained with high accuracy.

[0011] In addition, the neoantigen prediction device of the fourth invention is a neoantigen prediction device that, compared to any one of the first to third inventions, further includes a translation unit that refers to a DNA genetic code table, which is a correspondence table between character strings that make up a kmer and characters that make up an amino acid sequence, and acquires amino acid sequence information that specifies the amino acid sequence for each of the one or more kmers determined by the determination unit, and the position acquisition unit acquires position information for each of the one or more amino acid sequence information acquired by the translation unit using position information of a read corresponding to the kmer corresponding to the amino acid sequence information.

[0012] With this configuration, neoantigens can be predicted with high accuracy, and position information of neoantigens can be obtained with high accuracy.

[0013] In addition, the neoantigen prediction device of the fifth invention is a neoantigen prediction device in which, compared to any one of the first to fourth inventions, the position acquisition unit is equipped with a score acquisition means for acquiring a score regarding whether the amino acid sequence information is a neoantigen for each of one or more amino acid sequence information acquired by the translation unit, a neoantigen selection means for acquiring one or more amino acid sequence information whose score satisfies the neoantigen condition, and a position acquisition means for acquiring position information of each of the one or more amino acid sequence information acquired by the neoantigen selection means using position information of a lead corresponding to the kmer corresponding to the amino acid sequence information.

[0014] With this configuration, neoantigens can be predicted with higher accuracy, and position information of neoantigens can be obtained with higher accuracy. Effect of the Invention

[0015] According to the neoantigen prediction device of the present invention, neoantigens can be predicted with high accuracy, and position information of neoantigens can be obtained with high accuracy. [Brief description of the drawings]

[0016] [Figure 1] Block diagram of neoantigen prediction device A in embodiment 1 [Diagram 2] A flowchart explaining an example of the operation of the neoantigen prediction device A [Diagram 3] A flowchart illustrating an example of the mapping process. [Figure 4] A flowchart for explaining an example of the range selection process [Diagram 5] A flowchart illustrating an example of the determination process. [Figure 6] A flowchart illustrating an example of the kmer acquisition process. [Figure 7] A flowchart illustrating an example of the filtering process. [Figure 8] A flowchart illustrating an example of the counting process. [Figure 9]A flowchart for explaining an example of the number of times in the read acquisition process [Figure 10] A flowchart illustrating an example of the kmer selection process. [Figure 11] A flowchart illustrating an example of the information acquisition process [Figure 12] A flowchart illustrating an example of the translation process. [Figure 13] A flowchart illustrating an example of the neoantigen selection process [Figure 14] A flowchart illustrating an example of a position acquisition process [Figure 15] A flowchart for explaining an example of the configuration process [Figure 16] A diagram explaining the operation flow of the neoantigen prediction device A. [Figure 17] A diagram showing the DNA genetic code table [Figure 18] Figure showing an example of output of neoantigen information [Figure 19] Overview of the computer system [Figure 20] Block diagram of the computer system DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] Hereinafter, embodiments of the neoantigen prediction device and the like will be described with reference to the drawings. Note that components with the same reference numerals in the embodiments perform similar operations, and therefore repeated description may be omitted.

[0018] (Embodiment 1) In this embodiment, a neoantigen prediction device that acquires and outputs neoantigen information related to neoantigens from abnormal genome information is described. Note that genome information is genetic information on nucleic acids possessed by an organism. The genome information here is usually a collection of character strings corresponding to one or more chromosome identifiers. The character strings indicate base sequences. Abnormal genome information is genome information of unhealthy individuals such as cancer patients.

[0019] In this specification, information X being associated with information Y means that information Y can be obtained from information X, or information X can be obtained from information Y, and the method of association is not important. Information X and information Y may be linked, may exist in the same buffer, information X may be included in information Y, or information Y may be included in information X, etc.

[0020] Furthermore, in this specification, selecting information Z means obtaining information Z, obtaining a pointer to information Z, obtaining the ID of information Z, setting a flag for information Z, etc., as long as information Z can be accessed.

[0021] FIG. 1 is a block diagram of a neoantigen prediction device A in this embodiment. The neoantigen prediction device A may be a standalone device or a server. When the neoantigen prediction device A is a server, it may be, for example, a cloud server or an ASP server, but the type is not important. When the neoantigen prediction device A is a server, it receives information and instructions from a user terminal (not shown) and transmits processing results such as neoantigen information described below to the user terminal.

[0022] The neoantigen prediction device A includes a storage unit 1, a reception unit 2, a processing unit 3, and an output unit 4.

[0023] The storage unit 1 includes, for example, a reference genome sequence management unit 11, a normal database management unit 12, a code table management unit 13, and a range management unit 14. The reference genome sequence management unit 11, the normal database management unit 12, the code table management unit 13, and the range management unit 14 may be present in an external device other than the neoantigen prediction device A. The reception unit 2 includes, for example, a target reception unit 21, and a range reception unit 22. The processing unit 3 includes, for example, a mapping unit 31, a selection unit 32, a determination unit 33, a translation unit 34, a position acquisition unit 35, and a neoantigen acquisition unit 36. The determination unit 33 includes, for example, a kmer acquisition means 331, a determination means 332, a counting means 333, and a kmer selection means 334. The position acquisition unit 35 includes, for example, a score acquisition means 351, a neoantigen selection means 352, and a position acquisition means 353. The output unit 4 includes a neoantigen output unit 41.

[0024] Various types of information are stored in the storage unit 1 of the neoantigen prediction device A. The various types of information include, for example, reference genome sequence information described below, a normal database described below (hereinafter, appropriately referred to as a "normal DB"), a DNA genetic code table described below, and range information described below.

[0025] The reference genome sequence management unit 11 stores reference genome sequence information. The reference genome sequence information is information that stores the base sequence of each chromosome. For example, in the case of humans, the reference genome sequence information that is currently standard information is that published as GRCh38. The reference genome sequence information is, for example, a set of pairs of a chromosome identifier and the base sequence of a chromosome identified by the chromosome identifier. Note that the base sequence is, for example, a character string using "A", "C", "G", "T", and "N". The reference genome sequence information is genome information that is referred to by the mapping unit 31 described later. The reference genome sequence information is normal genome information. The normal genome information is genome information of a normal person. A normal person is a healthy person, for example, a person who does not suffer from a disease such as cancer. The reference genome sequence information is genome information of one or more normal people. The reference genome sequence information is information that is the source of acquiring position information of the reads that constitute the genome information.

[0026] A read is a part of genomic information. A read is usually a part of genomic information of a predetermined length. The predetermined length is a first length (hereinafter, appropriately referred to as "LL"). The first length is, for example, 150 characters, but is not limited thereto. A read is, for example, a base sequence of 150 characters.

[0027] The reference genome sequence information is, for example, a text file. The reference genome sequence information is, for example, a fasta file. For details about fasta files, see the web pages at the following URLs (https: / / biotech-lab.org / articles / 6389, https: / / ja.wikipedia.org / wiki / FASTA).

[0028] A normal database is stored in the normal database management unit 12. The normal database is a set of two or more kmers obtained from normal genome information. Note that a kmer may be a k-mer.

[0029] Normal genome information is, for example, a collection of genome information from one or more normal individuals. Normal genome information is, for example, fastq data or bam files. Fastq is text data of base sequences. Fastq is usually a file obtained for each case, but it may also be a file obtained from a normal individual. For more information on fastq, see, for example, the web pages (https: / / olvtools.com / documents / fastq, https: / / ja.wikipedia.org / wiki / Fastq).

[0030] A kmer is a portion of genome information with a predetermined length. The predetermined length in a kmer is a second length. The second length is equal to or less than the first length. The second length is, for example, 33 characters, but is not critical. A kmer is, for example, a 33-character base sequence. When the second length and the first length are the same, the kmer is a read.

[0031] The normal DB is, for example, information that has, for each kmer, the kmer and the number of occurrences in normal genome information. The normal DB is, for example, the "kmer hashmap (DB)" described later. It is.

[0032] The normal DB is preferably a relational database, but the data format is not important, and may be a CSV file, a TSV file, a text file, or the like.

[0033] The code table management unit 13 stores a DNA genetic code table. The DNA genetic code table is a correspondence table used when converting a kmer, which is a base sequence, into an amino acid sequence. The DNA genetic code table is a correspondence table between a character string constituting a kmer and a character constituting an amino acid sequence. The character string constituting a kmer is usually of fixed length. The character string constituting a kmer is, for example, three characters, but is not limited thereto. The character constituting an amino acid sequence is usually one character. The characters constituting an amino acid sequence are, for example, "F", "L", "I", "M", "V", "S", "P", "T", "A", "Y", "H", "Q", "N", "K", "D", "E", "C", "W", "R", "S", and "G". For the DNA genetic code table, refer to, for example, a web page (URL "https: / / www.nacalai.co.jp / information / trivia2 / 09.html").

[0034] The range management unit 14 stores one or more pieces of range information. The range information is information that specifies the range to be examined in the target genome information described later. The examination is a process for acquiring neoantigen information. Neoantigens are antigens produced by genetic mutations in cancer cells.

[0035] The range information is usually a region that is expected to be a neoantigen. The range information is, for example, a chromosome identifier, a first offset, and a second offset. The first offset is information that specifies the position of the start point in a character string of a chromosome identified by the chromosome identifier. The second offset is information that specifies the position of the end point in a character string of a chromosome identified by the chromosome identifier. The range information is, for example, information that specifies a character string in a range indicated by the first offset and the second offset in a character string of a chromosome identified by the chromosome identifier.

[0036] The range information is, for example, a chromosome identifier. In this case, the range indicated by the range information is the entire chromosome identified by the chromosome identifier.

[0037] The range information is, for example, a chromosome identifier and a first offset. In this case, the length of the range indicated by the range information is determined.

[0038] The reception unit 2 receives various information and instructions, such as target genome information, one or more reads in the target genome information, range information, and instructions for constructing a normal DB, which will be described later.

[0039] The normal DB configuration instruction is an instruction to configure a normal DB. Such a configuration instruction includes, for example, information on the location of the fastq file of normal genome information (for example, a URL).

[0040] The reception unit 2 receives various information and instructions from, for example, a user terminal (not shown). The reception unit 2 receives various information and instructions from, for example, an input means operated by a user.

[0041] Here, reception means acquiring various kinds of information and instructions, and is a concept that includes reception of information input from input devices such as a keyboard, mouse, or touch panel, reception of information transmitted via a wired or wireless communication line, reception of information read from a recording medium such as an optical disk, a magnetic disk, or a semiconductor memory, etc.

[0042] The means for inputting various information and instructions may be anything, such as a touch panel, a keyboard, a mouse, or a menu screen.

[0043] The subject receiving unit 21 receives one or more reads, which are a first length portion of the subject genome information. The subject genome information is genome information of a subject for predicting neoantigens. The subject genome information includes abnormal genome information. The abnormal genome information is usually genome information of a person having an abnormality. The person having an abnormality is, for example, a cancer patient.

[0044] The range receiving unit 22 receives target range information. The range receiving unit 22 acquires the target range information from, for example, the range management unit 14. The target range information is information that specifies the range to be inspected in the target genome information. The target range information is range information that specifies the target.

[0045] The processing unit 3 performs various types of processing. The various types of processing are, for example, processing performed by a mapping unit 31, a selection unit 32, a determination unit 33, a translation unit 34, a position acquisition unit 35, or a neoantigen acquisition unit 36.

[0046] The mapping unit 31 refers to the reference genome sequence information of the reference genome sequence management unit 11, and acquires position information that is information specifying the position of each of the one or more reads accepted by the target receiving unit 21. The position information is, for example, range information. That is, the position information has, for example, a chromosome identifier, a first offset, and a second offset. The position information has, for example, a chromosome identifier and a first offset. The position information has, for example, a chromosome identifier. The mapping unit 31 usually associates the acquired position information with the read for each of the one or more reads accepted by the target receiving unit 21. The process performed by the mapping unit 31 is called a mapping process.

[0047] For example, a bam file is generated as a result of the mapping process performed by the mapping unit 31. For example, a sam file is generated as a result of the mapping process performed by the mapping unit 31. Note that the data structure of the information generated by the mapping unit 31 is not limited.

[0048] A BAM file is a file that contains information that associates the acquired position information with the read for each of one or more reads. A BAM file is a SAM binary file. A SAM file is a text file obtained by mapping fastq. A SAM file is a tab-delimited file that contains information that links position information to sequence information (here, reads).

[0049] The selection unit 32 selects one or more leads corresponding to the position of the range specified by the range information received by the range receiving unit 22 from among the one or more leads received by the target receiving unit 21, using the position information of each of the one or more leads acquired by the mapping unit 31. The selection unit 32 can output neoantigen information at high speed. Note that it is preferable that the range information is information input based on the knowledge of a cancer treatment specialist, for example.

[0050] The determination unit 33 acquires two or more kmers that are portions of the second length (first length>=second length) from each of the one or more reads. The determination unit 33 acquires two or more kmers that are character strings of the second length (appropriately referred to as "kL") from each of the one or more reads, for example, by shifting each character by one. After acquiring two or more kmers, the determination unit 33 determines one or more kmers that do not exist in the normal DB from the two or more kmers.

[0051] The one or more leads from which the determining unit 33 obtains two or more kmers are preferably one or more leads selected by the selecting unit 32, but may be one or more leads accepted by the target accepting unit 21. When the one or more leads from which the determining unit 33 obtains two or more kmers are one or more leads accepted by the target accepting unit 21, the selecting unit 32 is not necessary.

[0052] The kmer acquisition means 331 included in the determination unit 33 acquires one or more kmers from each of one or more leads. For example, the kmer acquisition means 331 acquires 117 kmers, which are a character string of a second length (for example, 33 characters), from each of one or more leads (for example, 150 characters) while shifting each character by one.

[0053] The determining means 332 determines, from among one or more kmers acquired by the kmer acquiring means 331, one or more kmers that do not exist in the normal DB.

[0054] The counting means 333 acquires the number of occurrences of each of the one or more kmers in each of the one or more leads. The kmers for which the counting means 333 acquires the number of occurrences are preferably the one or more kmers determined by the determining means 332, but may be the one or more kmers acquired by the kmer acquiring means 331.

[0055] The kmer selection means 334 selects one or more kmers whose occurrence count satisfies the count condition from the one or more kmers determined by the determination means 332. The count condition is a condition related to the occurrence count. The count condition is a condition that is satisfied when the occurrence count is high. A kmer that satisfies the count condition is a kmer whose occurrence count is high enough to satisfy the count condition. The count condition is, for example, that the ranking of the occurrence count is in the top N or higher (N is a natural number equal to or greater than 1), or that the occurrence count is equal to or greater than a threshold value. The occurrence count here is usually the number of occurrences of the kmer in one or more leads after selection by the selection unit 32, but may also be the number of occurrences of the kmer in one or more leads before selection by the selection unit 32.

[0056] The translation unit 34 obtains amino acid sequence information that specifies the amino acid sequence for each of one or more kmers determined by the determination unit 33. When a kmer has, for example, 33 letters, the amino acid sequence information has, for example, 11 letters.

[0057] The translation unit 34 refers to, for example, the DNA genetic code table in the code table management unit 13 and obtains amino acid sequence information that specifies the amino acid sequence for each of the one or more kmers determined by the determination unit 33 .

[0058] The translation unit 34 performs machine learning prediction processing using, for example, each of the one or more kmers determined by the determination unit 33 and a learning model, and obtains amino acid sequence information corresponding to each of the one or more kmers.

[0059] The learning model is information configured by the learning process of machine learning and is information used for the prediction process of machine learning. The learning model here is information acquired by performing the learning process of machine learning using two or more teacher data having kmer and amino acid sequence information. The learning model may be called a learner, a classifier, a classification model, etc. The machine learning algorithm may be deep learning, random forest, decision tree, etc. In addition, for example, various machine learning functions such as TensorFlow (registered trademark) library, random forest module of R language, fastText, etc., and various existing libraries can be used for machine learning.

[0060] The position acquisition unit 35 acquires position information of the amino acid sequence information corresponding to each of one or more kmers determined by the determination unit 33, which is position information of the read corresponding to the kmer corresponding to the amino acid sequence information, using position information acquired by the mapping unit 31. The amino acid sequence information here is preferably information acquired by the translation unit 34, but may be a kmer. The position information of the amino acid sequence information is position information of the kmer that is the source of the amino acid sequence information.

[0061] The position acquisition unit 35 acquires, for example, position information of each of the one or more pieces of amino acid sequence information acquired by the translation unit 34, using position information of a read corresponding to a kmer corresponding to the amino acid sequence information.

[0062] The score acquisition means 351 included in the position acquisition unit 35 acquires a score for each of the one or more pieces of amino acid sequence information acquired by the translation unit 34. The score acquisition means 351 acquires a score, for example, by using information on the HLA type of the user. The HLA type information is, for example, information accepted by the acceptance unit 2 or information stored in the storage unit 1, but the acquisition route, etc., does not matter. The user here is a user corresponding to the target genome information. The score here is an evaluation value regarding whether the amino acid sequence information is a neoantigen. The score is, for example, the likelihood that the amino acid sequence information is a neoantigen, or the probability that the amino acid sequence information is a neoantigen, or the degree of neoantigen-likeness of the amino acid sequence information, or a predicted score of affinity to MHC. The score may also be an evaluation value regarding whether the amino acid sequence information is not a neoantigen. In other words, the score may be the likelihood that the amino acid sequence information is not a neoantigen, the probability that the amino acid sequence information is not a neoantigen, or the degree of neoantigen-likeness of the amino acid sequence information.

[0063] The score acquisition means 351 acquires a score by machine learning prediction processing using, for example, amino acid sequence information and a learning model. " is a module that performs MHC affinity prediction. The learning model here is, for example, information configured by a machine learning learning process using two or more pieces of teacher data having amino acid sequence information and a score. The learning model here is, for example, information configured by a machine learning learning process using two or more pieces of teacher data having amino acid sequence information and information indicating whether the amino acid sequence information is a neoantigen.

[0064] The machine learning algorithm here is preferably deep learning, but may be random forest, decision tree, SVM, etc. For machine learning, various machine learning functions such as the TensorFlow (registered trademark) library, the random forest module of the R language, fastText, TinySVM, etc., and various existing libraries can be used.

[0065] The neoantigen selection means 352 acquires one or more pieces of amino acid sequence information whose score satisfies the neoantigen condition. The neoantigen condition is a condition for determining that the amino acid sequence information is a neoantigen. The neoantigen condition is, for example, that the score acquired by the score acquisition means 351 is equal to or greater than a threshold value, or that the score acquired by the score acquisition means 351 is in the top N or higher (N is a natural number equal to or greater than 1).

[0066] The position acquisition means 353 acquires position information of one or more pieces of amino acid sequence information acquired by the neoantigen selection means 352, using position information of a lead corresponding to a kmer corresponding to the amino acid sequence information. The position acquisition means 353 may acquire position information paired with a kmer corresponding to one or more pieces of amino acid sequence information acquired by the neoantigen selection means 352. The position information acquired by the position acquisition means 353 is, for example, a chromosome identifier specified by the position information of the lead and range information indicating the range of the kmer within the range indicated by the range information.

[0067] The neoantigen acquisition unit 36 ​​acquires neoantigen information, which is information that associates the position information acquired by the position acquisition unit 35 with one or more pieces of amino acid sequence information. The neoantigen information includes, for example, amino acid sequence information predicted to be a neoantigen and position information of a kmer corresponding to the amino acid sequence information.

[0068] The output unit 4 outputs various types of information, such as neoantigen information.

[0069] Here, output is a concept that includes display on a display, projection using a projector, printing on a printer, sound output, transmission to an external device, storage on a recording medium, and handing over the processing results to other processing devices or other programs, etc.

[0070] The neoantigen output unit 41 outputs the neoantigen information acquired by the neoantigen acquisition unit 36.

[0071] The neoantigen output unit 41 transmits the neoantigen information to, for example, a user terminal (not shown). The neoantigen output unit 41 displays the neoantigen information on a display, for example.

[0072] The storage unit 1, the reference genome sequence management unit 11, the normal database management unit 12, the code table management unit 13, and the range management unit 14 are preferably non-volatile recording media, but may also be realized as volatile recording media.

[0073] There is no restriction on the process by which information is stored in the storage unit 1, etc. For example, information may be stored in the storage unit 1, etc. via a recording medium, information transmitted via a communication line, etc. may be stored in the storage unit 1, etc., or information inputted via an input device may be stored in the storage unit 1, etc.

[0074] The reception unit 2, the target reception unit 21, and the range reception unit 22 may be realized, for example, by a device driver for an input means such as a touch panel or a keyboard, control software for a menu screen, etc. The reception unit 2, etc. are realized, for example, by a wireless or wired communication means.

[0075] The processing unit 3, mapping unit 31, selection unit 32, determination unit 33, translation unit 34, position acquisition unit 35, neoantigen acquisition unit 36, kmer acquisition means 331, determination means 332, counting means 333, kmer selection means 334, score acquisition means 351, neoantigen selection means 352, and position acquisition means 353 can usually be realized by a processor, memory, etc. The processing procedure of the processing unit 3, etc. is usually realized by software, and the software is recorded in a recording medium such as a ROM. However, it may also be realized by hardware (dedicated circuit). The processor may be a CPU, MPU, GPU, etc., and the type is not important.

[0076] The output unit 4 and the neoantigen output unit 41 can be realized, for example, by driver software for an output device, or a driver software for an output device and an output device, etc. The output unit 4 and the neoantigen output unit 41 are realized, for example, by wireless or wired communication means.

[0077] The output unit 4 etc. may or may not include output devices such as a display and a speaker.

[0078] Next, an example of the operation of the neoantigen prediction device A will be described with reference to the flowchart of FIG.

[0079] (Step S201) The target receiving unit 21 judges whether or not one or more reads acquired from the target genome information have been received. If one or more reads have been received, the process proceeds to step S202. If one or more reads have not been received, the process proceeds to step S208. The target genome information is the genome information of the target from which neoantigen information is to be acquired.

[0080] (Step S202) The mapping unit 31 performs a mapping process. An example of the mapping process will be described with reference to the flowchart in FIG. 3. The mapping process is a process of mapping one or more received reads. The mapping process is a process of acquiring position information that identifies the position of the read in the genome information and associating it with the read.

[0081] (Step S203) The selection unit 32 performs a range selection process. An example of the range selection process will be described with reference to the flowchart of Fig. 4. The range selection process is a process of narrowing down the reads from which neoantigen information is to be obtained from one or more received reads using the received range information.

[0082] (Step S204) The determination unit 33 performs a determination process. An example of the determination process will be described with reference to the flowchart in FIG. 5. The determination process is a process for determining the kmer of a neoantigen. The kmer here may be the kmer of a candidate neoantigen.

[0083] (Step S205) The translation unit 34 and the like perform an information acquisition process. An example of the information acquisition process will be described with reference to the flowchart of Fig. 11. The information acquisition process is a process for acquiring amino acid sequence information and position information constituting the neoantigen information to be output.

[0084] (Step S206) The neoantigen acquisition unit 36 ​​constructs one or more pieces of neoantigen information having the amino acid sequence information and position information acquired in step S205.

[0085] (Step S207) The neoantigen output unit 41 outputs one or more pieces of neoantigen information constructed in step S206. Return to step S201.

[0086] (Step S208) The reception unit 2 judges whether or not an instruction to configure a normal database has been received. If the instruction to configure the normal database has been received, the process proceeds to step S209, and if the instruction to configure the normal database has not been received, the process returns to step S201.

[0087] (Step S209) The processing unit 3 performs a configuration process. Return to step S201. An example of the configuration process will be described with reference to the flowchart in Fig. 15. Note that the configuration process is a process of creating a normal database from information on one or more normal genomes.

[0088] In the flowchart of FIG. 2, the range selection process in step S203 may be omitted.

[0089] In the flowchart of FIG. 2, the process ends when the power is turned off or an interrupt occurs to end the process.

[0090] Next, an example of the mapping process in step S202 will be described with reference to the flowchart in FIG.

[0091] (Step S301) The mapping unit 31 assigns 1 to a counter i.

[0092] (Step S302) The mapping unit 31 judges whether or not the i-th lead exists among the one or more leads accepted in step S201. If the i-th lead exists, the process proceeds to step S303, and if the i-th lead does not exist, the process returns to the upper process.

[0093] (Step S303) The mapping unit 31 acquires the i-th lead from the one or more leads accepted in step S201.

[0094] (Step S304) The mapping unit 31 assigns 1 to a counter j.

[0095] (Step S305) The mapping unit 31 judges whether or not genome information corresponding to the j-th chromosome identifier is present in the normal database. If genome information corresponding to the j-th chromosome identifier is present, the process proceeds to step S306, and if not, the process proceeds to step S314.

[0096] (Step S306) The mapping unit 31 assigns 1 to a counter k.

[0097] (Step S307) The mapping unit 31 judges whether or not the (k+LL-1)th character exists in the genome information corresponding to the jth chromosome identifier. If the (k+LL-1)th character exists, the process proceeds to step S308. If the (k+LL-1)th character does not exist, the process proceeds to step S313. Here, LL is the first length, which is the length of the read.

[0098] (Step S308) The mapping unit 31 acquires a reference character string that is a character string of a first length "LL" from the kth character in the genome information corresponding to the jth chromosome identifier. The reference character string is a character string from the kth character to the (k+LL-1)th character in the genome information corresponding to the jth chromosome identifier.

[0099] (Step S309) The mapping unit 31 acquires the similarity between the i-th lead and the reference string acquired in step S308. Any algorithm can be used to acquire the similarity between the two strings (the i-th lead and the reference string). For example, the similarity between the two strings may be the number of matching characters in the two strings, or the cosine similarity between two vectors in which each character in the strings is replaced with a number.

[0100] (Step S310) The mapping unit 31 judges whether the similarity acquired in step S309 is large enough to satisfy the mapping condition. If the mapping condition is satisfied, the process proceeds to step S311, and if the mapping condition is not satisfied, the process proceeds to step S312. Note that the mapping condition is, for example, that the similarity is equal to or greater than a threshold value.

[0101] (Step S311) The mapping unit 31 acquires position information. The position information is, for example, the jth chromosome identifier, and the offset of the first character of the string and the offset of the last character (here, for example, (k, k+LL-1)). The position information is, for example, the jth chromosome identifier, the offset of the first character of the string, and the first length (here, "LL"), or the jth chromosome identifier and the offset of the first character of the string. When the position information is only the offset of the first character of the string, the range indicated by the position information is determined using the first length (LL) stored in the storage unit 1. The position information is, for example, the jth chromosome identifier, the offset of the first character of the string in the first part, the offset of the last character of the first part, and the offset of the first character of the string in the second part, and the offset of the last character of the second part. That is, the object to be mapped may not be continuous and may be divided into two or more parts.

[0102] (Step S312) The mapping unit 31 increments the counter k by 1. The process returns to step S307.

[0103] (Step S313) The mapping unit 31 increments the counter j by 1. The process returns to step S305.

[0104] (Step S314) The mapping unit 31 increments the counter i by 1. The process returns to step S302.

[0105] In the flowchart of Fig. 3, when two or more pieces of position information are obtained for one lead, the position information of the reference string with the highest similarity may be adopted. In other words, an algorithm may be used in which only one piece of position information is adopted for one lead, or an algorithm may be used in which two or more pieces of position information can be adopted. Furthermore, the flowchart in FIG. 3 is one example of a process for acquiring lead position information, and it goes without saying that other methods may be used.

[0106] Next, an example of the range selection process in step S203 will be described with reference to the flowchart in FIG.

[0107] (Step S401) Range receiving unit 22 acquires one or more pieces of target range information. Note that the one or more pieces of target range information may be information received from a user or information stored in storage unit 1 in advance.

[0108] (Step S402) The selection unit 32 assigns 1 to a counter i.

[0109] (Step S403) The selection unit 32 judges whether or not the i-th lead for which the mapping process was performed in step S202 exists. If the i-th lead exists, the process proceeds to step S404, and if not, the process returns to the upper process.

[0110] (Step S404) The selection unit 32 acquires one or more pieces of position information corresponding to the i-th lead.

[0111] (Step S405) The selection unit 32 assigns 1 to a counter j.

[0112] (Step S406) The selection unit 32 judges whether or not the j-th target range information exists among the one or more pieces of target range information acquired in step S401. If the j-th target range information exists, the process proceeds to step S407, and if not, the process proceeds to step S413.

[0113] (Step S407) The selection unit 32 assigns 1 to a counter k.

[0114] (Step S408) The selection unit 32 judges whether or not the k-th location information exists among the one or more pieces of location information acquired in step S404. If the k-th location information exists, the process proceeds to step S409, and if not, the process proceeds to step S413.

[0115] (Step S409) The selection unit 32 judges whether the kth location information and the jth target range information corresponding to the ith lead satisfy the range condition. If the range condition is satisfied, the process proceeds to step S410, and if the range condition is not satisfied, the process proceeds to step S412. The range condition is a condition regarding the range of the two location information. The range condition is, for example, that the two location information match, that the ranges specified by the two location information overlap, that the range specified by one location information is included in the range specified by the other location information, or that the range specified by one location information completely includes the range specified by the other location information.

[0116] (Step S410) The selection unit 32 selects the i-th lead.

[0117] (Step S411) The selection unit 32 increments the counter i by 1. The process returns to step S403.

[0118] (Step S412) The selection unit 32 increments the counter k by 1. The process returns to step S408.

[0119] (Step S413) The selection unit 32 increments the counter j by 1. The process returns to step S406.

[0120] (Step S414) The selection unit 32 increments the counter i by 1. The process returns to step S403.

[0121] Next, an example of the determination process in step S204 will be described with reference to the flowchart in FIG.

[0122] (Step S501) The kmer acquisition means 331 constituting the determination unit 33 performs a kmer acquisition process. An example of the kmer acquisition process will be described with reference to the flowchart in Fig. 6. The kmer acquisition process is a process for acquiring one or more kmers from one or more leads. The kmer acquisition process is usually a process for acquiring two or more kmers from one or more leads.

[0123] (Step S502) The kmer acquisition means 331 performs unique processing on the one or more kmers acquired in step S501.

[0124] As a result of the unique processing, there will be no duplicate kmers in the buffer (not shown). If two or more kmers are the same among the kmers acquired in step S501, the position information of each kmer will be associated with the kmer after the unique processing, and the number of occurrences will be added up and associated with the kmer after the unique processing.

[0125] (Step S503) The determining means 332 performs a filtering process. An example of the filtering process will be described with reference to the flowchart in Fig. 7. The filtering process is a process of determining a kmer that does not exist in the normal database from two or more acquired kmers.

[0126] (Step S504) The counting means 333 performs a counting process. An example of the counting process will be described with reference to the flowchart in Fig. 8. The counting process is a process of acquiring the number of occurrences of one or more selected kmers in one or more leads.

[0127] (Step S505) The kmer selection means 334 performs a kmer selection process. The process returns to the upper process. An example of the kmer selection process will be described with reference to the flowchart in FIG. 10. The kmer selection process is a process of selecting one or more kmers for acquiring neoantigen information using the number of occurrences.

[0128] Next, an example of the kmer acquisition process in step S501 will be described with reference to the flowchart in FIG.

[0129] (Step S601) The kmer acquiring means 331 assigns 1 to a counter i.

[0130] (Step S602) The kmer acquisition means 331 judges whether or not the i-th lead exists among the one or more leads selected in step S203. If the i-th lead exists, the process proceeds to step S603, and if not, the process proceeds to step S612.

[0131] (Step S603) The kmer acquisition means 331 acquires the i-th lead from among the one or more leads selected in step S203.

[0132] (Step S604) The kmer acquisition means 331 acquires the second length (kL). Note that the second length (kL) is determined in advance.

[0133] (Step S605) The kmer acquiring means 331 assigns 1 to the counter j.

[0134] (Step S606) The kmer acquisition means 331 judges whether or not the (j+kL-1)th character exists in the i-th lead. If the (j+kL-1)th character exists, the process proceeds to step S607. If not, the process proceeds to step S611.

[0135] (Step S607) The kmer acquisition means 331 acquires the j-th kmer, which is the character string from the j-th to the (j+kL-1)-th characters (character string of the second length), from the i-th lead.

[0136] (Step S608) The kmer acquisition means 331 acquires the position information of the j-th kmer. The position information of the j-th kmer is acquired using the chromosome identifier corresponding to the i-th read, the range information, "j", and "j+kL-1". When the range information is (x, y), the position information of the j-th kmer is, for example, the chromosome identifier corresponding to the i-th read and (x+j-1, x+j+kL-2). In the range information, x is the first offset, and y is the second offset.

[0137] (Step S609) The kmer acquisition means 331 associates the j-th kmer acquired in step S607 with the location information acquired in step S608, and accumulates them in a buffer (not shown).

[0138] (Step S610) The kmer acquiring means 331 increments the counter j by 1. The process returns to step S606.

[0139] (Step S611) The kmer acquiring means 331 increments the counter i by 1. The process returns to step S602.

[0140] Next, an example of the filtering process in step S503 will be described with reference to the flowchart in FIG.

[0141] (Step S701) The determining means 332 assigns 1 to a counter i.

[0142] (Step S702) The determination means 332 judges whether or not the i-th kmer exists among one or more kmers resulting from the unique processing in step S502. If the i-th kmer exists, the process proceeds to step S703, and if not, the process returns to the upper level process.

[0143] (Step S703) The determining means 332 acquires the i-th kmer from among one or more kmers resulting from the unique processing in step S502.

[0144] (Step S704) The determining means 332 assigns 1 to the counter j.

[0145] (Step S705) The decision means 332 judges whether or not the j-th kmer exists in the normal DB. If the j-th kmer exists, the process proceeds to step S706, and if not, the process proceeds to step S710.

[0146] (Step S706) The determining means 332 acquires the j-th kmer in the normal DB.

[0147] (Step S707) The determination means 332 judges whether the i-th kmer acquired in step S703 and the j-th kmer acquired in step S706 match. If they match, the process proceeds to step S708, and if they do not match, the process proceeds to step S709.

[0148] (Step S708) The determination means 332 increments the counter i by 1. The process returns to step S702.

[0149] (Step S709) The determination means 332 increments the counter j by 1. The process returns to step S705.

[0150] (Step S710) The determining means 332 temporarily stores the i-th kmer acquired in step S703 in a buffer (not shown).

[0151] (Step S711) The determination means 332 increments the counter i by 1. The process returns to step S702.

[0152] Next, an example of the counting process in step S504 will be described with reference to the flowchart in FIG.

[0153] (Step S801) The counting means 333 assigns 1 to the counter i.

[0154] (Step S802) The counting means 333 judges whether or not the i-th kmer is present among one or more kmers temporarily stored in a buffer (not shown) in step S503. If the i-th kmer is present, the process proceeds to step S803, and if not, the process returns to the upper process.

[0155] (Step S803) The counting means 333 assigns 1 to the counter j.

[0156] (Step S804) The counting means 333 judges whether or not the j-th lead exists among the one or more leads to be inspected. If the j-th lead exists, the process proceeds to step S805, and if not, the process proceeds to step S808.

[0157] (Step S805) The counting means 333 acquires the number of times that the i-th kmer appears in the j-th lead. An example of such in-lead number acquisition processing will be described with reference to the flowchart of FIG.

[0158] (Step S806) The counting means 333 adds the number of occurrences acquired in step S805 to the number of occurrences. Note that the initial value of the number of occurrences is "0".

[0159] (Step S807) The counting means 333 increments the counter j by 1. The process returns to step S804.

[0160] (Step S808) The counting means 333 associates the number of occurrences acquired in step S806 with the i-th kmer.

[0161] (Step S809) The counting means 333 increments the counter i by 1. The process returns to step S802.

[0162] Next, an example of the in-read count acquisition process in step S805 will be described with reference to the flowchart in FIG.

[0163] (Step S901) The counting means 333 assigns "0" to the variable "number of times".

[0164] (Step S902) The counting means 333 acquires a target lead. Note that the target lead is the j-th lead of step S804.

[0165] (Step S903) The counting means 333 assigns 1 to the counter i.

[0166] (Step S904) The counting means 333 judges whether or not the (i+kL-1)th character exists in the lead acquired in step S902. If the character exists, the process proceeds to step S905, and if not, the process returns to the upper process.

[0167] (Step S905) The counting means 333 acquires a character string of a second length (kL) from the i-th character in the lead acquired in step S902. This character string is kmer.

[0168] (Step S906) The counting means 333 judges whether or not the i-th kmer in step S802 matches the kmer acquired in step S905. If the two kmers match, the process proceeds to step S907, and if they do not match, the process proceeds to step S908.

[0169] (Step S907) The counting means 333 adds 1 to the variable "number of times."

[0170] (Step S908) The counting means 333 increments the counter i by 1. The process returns to step S904.

[0171] Next, an example of the kmer selection process in step S505 will be described with reference to the flowchart in FIG.

[0172] (Step S1001) The kmer selection means 334 sorts the kmers in descending order using the number of occurrences as a key.

[0173] (Step S1002) The kmer selection means 334 assigns 1 to a counter i.

[0174] (Step S1003) The kmer selection means 334 judges whether or not the i-th kmer is present among the two or more kmers sorted in step S1001. If the i-th kmer is present, the process proceeds to step S1004, and if not, the process returns to the upper process.

[0175] (Step S1004) The kmer selection means 334 acquires the number of occurrences paired with the i-th kmer.

[0176] (Step S1005) The kmer selection means 334 judges whether the occurrence count satisfies the count condition. If the occurrence count satisfies the count condition, proceed to step S1005; if the occurrence count does not satisfy the count condition, return to higher-level processing. The count condition is information for selecting a kmer. The count condition is a condition related to the occurrence count of a kmer. The count condition is, for example, that the occurrence count is within the top N (N is a natural number equal to or greater than 1) and that the occurrence count is equal to or greater than a threshold value.

[0177] (Step S1005) The kmer selection means 334 judges whether or not the occurrence count satisfies the count condition. If the occurrence count condition is met, the process proceeds to step S1006, and if the occurrence count condition is not met, the process returns to the upper process.

[0178] (Step S1006) The kmer selection means 334 temporarily stores the i-th kmer in a buffer (not shown).

[0179] (Step S1007) The kmer selection means 334 increments the counter i by 1. The process returns to step S1003.

[0180] Next, an example of the information acquisition process in step S205 will be described with reference to the flowchart in FIG.

[0181] (Step S1101) The translation unit 34 performs a translation process. An example of the translation process will be described with reference to the flowchart in Fig. 12. The translation process is a process for converting a kmer into amino acid sequence information.

[0182] (Step S1102) The position acquisition unit 35 selects a neoantigen from one or more pieces of amino acid sequence information acquired in step S1101. An example of such a neoantigen selection process will be described with reference to the flowchart of FIG.

[0183] (Step S1103) The position acquisition unit 35 acquires the position information of one or more pieces of amino acid sequence information selected in step S1102. The process returns to the upper level process. An example of such position acquisition process will be described with reference to the flowchart of FIG.

[0184] Next, an example of the translation process in step S1101 will be described with reference to the flowchart in FIG.

[0185] (Step S1201) The translation unit 34 assigns 1 to a counter i.

[0186] (Step S1202) The translation unit 34 judges whether or not the i-th kmer is present among the one or more kmers determined in step S202. If the i-th kmer is present, the process proceeds to step S1203, and if not, the process returns to the upper process.

[0187] (Step S1203) The translation unit 34 acquires the i-th kmer from the one or more kmers determined in step S202.

[0188] (Step S1204) The translation unit 34 assigns 1 to a counter j.

[0189] (Step S1205) The translation unit 34 judges whether or not the jth character string of n characters exists in the i-th kmer acquired in step S1203. If the jth character string of n characters exists, the process proceeds to step S1206, and if not, the process proceeds to step S1209. Note that n is the length of the character string in the kmer in the correspondence between the character string in the kmer and the characters constituting the amino acid sequence information, and is usually "3".

[0190] (Step S1206) The translation unit 34 acquires a character string including the jth character through the (j+n-1)th character in the i-th kmer.

[0191] (Step S1207) The translation unit 34 obtains the characters of the amino acid sequence corresponding to the character string obtained in step S1206. Next, the translation unit 34 adds the characters to a buffer in which amino acid sequence information is stored. Note that the translation unit 34 obtains the characters of the amino acid sequence from, for example, a DNA genetic code table.

[0192] (Step S1208) The translation unit 34 increments the counter j by n (for example, "3"), and then returns to step S1205.

[0193] (Step S1209) The translation unit 34 increments the counter i by 1. The process returns to step S1202.

[0194] Next, an example of the neoantigen selection process in step S1102 will be described with reference to the flowchart in FIG.

[0195] (Step S1301) The score obtaining means 351 assigns 1 to a counter i.

[0196] (Step S1302) The score acquiring means 351 acquires the i-th amino acid sequence information from the buffer in which the amino acid sequence information is stored. Note that this buffer is the buffer in step S1207.

[0197] (Step S1303) The score acquisition means 351 acquires the score of the i-th amino acid sequence information. For example, the score acquisition means 351 provides the i-th amino acid sequence information to a module that predicts MHC affinity called "pVACbind," executes the module, and acquires the score.

[0198] (Step S1304) The neoantigen selection means 352 judges whether or not the score acquired in step S1303 satisfies the neoantigen condition. If the neoantigen condition is satisfied, the process proceeds to step S1305, and if not, the process proceeds to step S1306. The neoantigen condition is, for example, that the score is equal to or greater than a threshold value.

[0199] (Step S1305) The neoantigen selection means 352 temporarily stores the i-th amino acid sequence information in a buffer (not shown). The amino acid sequence information stored in the buffer is to be output.

[0200] (Step S1306) The score obtaining means 351 increments the counter i by 1. The process returns to step S1302.

[0201] Next, an example of the position acquisition process in step S1103 will be described with reference to the flowchart in FIG.

[0202] (Step S1401) The position obtaining means 353 assigns 1 to a counter i.

[0203] (Step S1402) Position acquisition means 353 judges whether or not the i-th amino acid sequence information exists in a buffer (not shown). If the i-th amino acid sequence information exists, the process proceeds to step S1403, and if not, the process returns to the upper process. Note that this buffer (not shown) is the buffer in step S1305.

[0204] (Step S1403) The position obtaining means 353 determines a kmer corresponding to the i-th amino acid sequence information. The kmer corresponding to the i-th amino acid sequence information is the kmer that is the source of the i-th amino acid sequence information (before translation processing).

[0205] (Step S1404) The position acquiring means 353 acquires position information paired with the kmer determined in step S1403. The position acquiring means 353 may acquire position information paired with the kmer using position information of the lead corresponding to the kmer determined in step S1403.

[0206] (Step S1405) The position acquiring means 353 associates the position information acquired in step S1404 with the i-th amino acid sequence information.

[0207] (Step S1406) The position obtaining unit 353 increments the counter i by 1. The process returns to step S1402.

[0208] Next, an example of the configuration process in step S209 will be described with reference to the flowchart in FIG.

[0209] (Step S1501) The kmer acquisition means 331 performs a process of acquiring kmers from two or more reads acquired from normal genome information. The kmer acquisition process is the process described using the flowchart in Fig. 6. The normal genome information is, for example, a fastq file based on genome information of a normal person.

[0210] (Step S1502) The counting means 333 acquires the number of occurrences of one or more kmers acquired in step S1501. The process returns to the upper level process. Note that this counting process is the process described using the flowchart in Fig. 8. The number of occurrences here is the number of occurrences of a kmer in two or more reads acquired from normal genome information.

[0211] Hereinafter, a specific operation flow of the neoantigen prediction device A in this embodiment will be described with reference to FIG.

[0212] Assume now that the code table management unit 13 of the neoantigen prediction device A stores the DNA genetic code table shown in FIG. 17. The DNA genetic code table is a table that manages two or more records having an "ID," "kmer string," and "amino acid sequence character." The "ID" is information that identifies a record. The "kmer string" is a string that makes up the kmer, and is three characters in this case. The "amino acid sequence character" is one character that makes up the amino acid sequence.

[0213] In the above situation, it is assumed that the user inputs an instruction to construct a normal database to the neoantigen prediction device A. Then, the reception unit 2 of the neoantigen prediction device A receives the instruction to construct a normal database. Next, the kmer acquisition means 331 of the processing unit 3 performs a process to acquire kmers from two or more reads acquired from the normal genome information. The kmer acquisition process is, for example, the process described using the flowchart of FIG. 6. The normal genome information is 1601, which is one or two or more files (R1, R2) of fastq in this case. Next, the counting means 333 of the processing unit 3 performs a counting process to acquire the number of occurrences of one or more kmers acquired by the kmer acquisition means 331. The counting process is, for example, the process described using the flowchart of FIG. 8. The module that performs the above kmer acquisition process and counting process is the "canonical k-mer count" 1602.

[0214] As a result of the above processing, the processing unit 3 creates a normal DB that includes two or more pairs of a kmer and the number of times the kmer appears in the normal genome information, and stores the normal DB in the normal database management unit 12. Note that the read here is a character string of 150 characters. The kmer here is a character string of 33 characters. The normal DB here is "kmer hashmap(DB)" 1603. The normal DB has been constructed by the above processing.

[0215] Next, assume that target genome information acquired from genome information of a cancer patient is stored in storage unit 1. The target genome information here is fastq files (R1, R2). The target genome information is 1604. Then, in response to an instruction from a user, target reception unit 21 acquires the fastq files (R1, R2), which are the target genome information, from storage unit 1.

[0216] Next, the mapping unit 31 performs the mapping process described using the flowchart of FIG. 3. That is, the mapping unit 31 refers to the reference genome sequence information and acquires position information that is information specifying the position of each of one or more reads contained in the target genome information 1604. Here, the reference genome sequence information is a fasta file. Furthermore, as a result of the mapping process, position information is associated with each of one or more reads. As a result, a bam file 1606 is created. The mapping process performed by the mapping unit 31 is performed by a module "mapping" 1607.

[0217] Next, the selection unit 32 and the like perform the range selection process described using the flowchart of FIG. 4. That is, the range reception unit 22 acquires one or more pieces of range information from the range management unit 14 shown in 1608. Next, the selection unit 32 uses position information corresponding to each of the one or more leads for which the mapping process has been performed to determine whether or not each of the one or more leads satisfies a range condition for the one or more pieces of range information from the range management unit 14. Next, the selection unit 32 acquires one or more leads that satisfy the range condition. The module that performs such range selection process is "extract nc-region" 1609. Also, as a result of the range selection process, a bam file 1610 having each of the one or more selected leads and position information is output.

[0218] Next, the determination unit 33 performs the determination process by the process described using the flowchart of FIG. 5. First, the kmer acquisition means 331 of the determination unit 33 acquires a large number of kmers from one or more reads of the bam file 1610. The kmer acquisition process has been described using the flowchart of FIG. 6. The counting means 333 performs the counting process described using the flowchart of FIG. 8 for each kmer acquired by the kmer acquisition process. The determination unit 33 acquires position information of each kmer acquired by the kmer acquisition process by using position information of the read that is the source of each kmer. As a result, information (1611) is constructed that includes a large number of kmers, position information paired with each kmer, and the number of occurrences of the kmer. The above process of the determination unit 33 is performed by the module "+standard convert k-mer count" 1612.

[0219] Next, the determining means 332 of the determining unit 33 performs the filtering process described using the flowchart of FIG. 7. That is, the determining means 332 refers to the normal database 1603, and performs a process of deleting kmers in the information (1611) that exist in the normal database 1603 and leaving kmers that do not exist in the normal database 1603. This process is performed by the module "subtract" 1613. By the process of the module "subtract" 1613, the data amount of the information (1611) is significantly reduced, and information (1614) having position information and occurrence counts paired with each kmer, which is small information, is formed. Note that the module "subtract" 1613 may store information (1614) having kmers and position information that do not exist in the normal database 1603 separately while leaving the information (1611).

[0220] Next, the kmer selection means 334 performs a process of leaving only the top n kmers (n is a natural number equal to or greater than 1) in terms of frequency of occurrence, for example, according to the operation of the flowchart of the kmer selection process in Fig. 10. This process is performed by a module "top n-th kmer assemble" 1615. Note that the process of leaving only the top n kmers in terms of frequency of occurrence may be a process of deleting kmers lower than the top n from the information (1611), or a process of leaving the information (1611) and separately accumulating the top n kmers in the information (1611). Note that, for example, each of the top n kmers is associated with location information.

[0221] Next, the translation unit 34 performs the translation process described using the flowchart of FIG. 12 for each of the top n kmers acquired by the module "top n-th kmer assemble" 1615. That is, the translation unit 34 refers to the DNA genetic code table (FIG. 17), converts a 3-letter kmer character string in each kmer into an amino acid sequence character string, and constructs a character string in which the 3-letter kmer character string is replaced with the amino acid sequence character string. Such a character string is amino acid sequence information, and is usually 11 characters. As a result of such translation process, one or more pairs having amino acid sequence information and position information are constructed. Note that such translation process is performed by the module "translate" 1616. Furthermore, the translation unit 34 acquires, for example, amino acid sequence information "FFL···V" from the kmer "TTTTTCTTA···GTT".

[0222] Next, the score acquisition means 351 acquires a score for each of the one or more pieces of amino acid sequence information that are the results of the translation process. That is, the score acquisition means 351 provides each of the one or more pieces of amino acid sequence information to the module "pVACbind" and acquires a score for each piece of amino acid sequence information. This module is 1617.

[0223] Next, the neoantigen selection means 352 judges whether the score of each piece of amino acid sequence information satisfies the neoantigen condition. Next, one or more pieces of amino acid sequence information that satisfy the neoantigen condition are acquired. The neoantigen condition is, for example, that the score is equal to or greater than a threshold value. The position acquisition means 353 acquires position information of each piece of amino acid sequence information that satisfies the neoantigen condition and associates it with the amino acid sequence information. The position information may be associated with the amino acid sequence information, or may be acquired from the bam file 1610, from the position information of the read corresponding to the kmer corresponding to the amino acid sequence information. The processing performed by the neoantigen selection means 352 and the position acquisition means 353 is performed by the module "pVACbind Modify" 1618.

[0224] Next, the neoantigen acquisition unit 36 ​​acquires neoantigen information using the output of the module “pVACbind Modify” 1618. Such neoantigen information includes, for example, one or more pieces of amino acid sequence information, position information of each piece of amino acid sequence information, and a score of each piece of amino acid sequence information.

[0225] Next, the neoantigen output unit 41 outputs the neoantigen information. The neoantigen information is 1619. An output example of the neoantigen information is shown in FIG.

[0226] In Figure 18, "Mutation" is the kmer ID. "HLA Allele" is the HLA type of the target user. "Sub-peptide Position" is the start position of the peptide in the kmer. "Epitope Seq" is the target amino acid sequence (neoantigen information). "Median Score" is the median of the score set obtained by prediction using multiple machine learning algorithms. "Best Score" is the best score among the score set obtained by prediction using multiple machine learning algorithms. "Locus" is the genomic coordinate (location information) of the Epitope Seq. "max_coverage" is the maximum number of reads on the genomic coordinates of the Epitope Seq. "Feature" is the feature on the genomic coordinates of the Epitope Seq.

[0227] As described above, according to the present embodiment, it is possible to predict neoantigens with high accuracy, and to obtain position information of neoantigens with high accuracy. In addition, in the present embodiment, since it is possible to obtain position information of neoantigens, it is possible to view genetic information surrounding the neoantigens, which is useful for diagnosis.

[0228] In addition, in this embodiment, the neoantigen prediction device A may not have the range management unit 14, the range acceptance unit 22, the selection unit 32, etc. Also, the storage unit 1 may be provided in an external device. In such a case, the neoantigen prediction device A includes a target acceptance unit 21 that accepts one or more reads that are a first length portion of target genome information including abnormal genome information, a mapping unit 31 that is normal genome information and refers to reference genome sequence information from which position information of the reads constituting the genome information is obtained, and acquires position information that is information that identifies the position of each of the one or more reads, and a positive storage unit that acquires two or more kmers that are a second length (first length >= second length) from each of the one or more reads, and stores the two or more kmers acquired from the normal genome information among the two or more kmers. The system comprises a determination unit 33 that determines one or more kmers that are not normally present in the database, a position acquisition unit 35 that acquires position information of amino acid sequence information corresponding to each of the one or more kmers determined by the determination unit 33, which is position information of the lead corresponding to the kmer corresponding to the amino acid sequence information, using the position information acquired by the mapping unit 31, a neoantigen acquisition unit 36 ​​that acquires neoantigen information which is information that associates the position information acquired by the position acquisition unit 35 with each of the one or more amino acid sequence information, and a neoantigen output unit 41 that outputs the neoantigen information.

[0229] Furthermore, the processing in this embodiment may be realized by software. This software may be distributed by software download or the like. This software may be recorded on a recording medium such as a CD-ROM and distributed. This also applies to other embodiments in this specification. The software that realizes the neoantigen prediction device A in this embodiment is a program such as the following. In other words, this program causes a computer to function as a target receiving unit that receives one or more reads that are a first length portion of target genome information including abnormal genome information; a mapping unit that refers to reference genome sequence information that is normal genome information and from which position information of the reads that constitute the genome information is obtained, and acquires position information that is information that identifies the position of each of the one or more reads; a determination unit that acquires two or more kmers that are a second length (first length >= second length) portion from each of the one or more reads, and determines one or more kmers among the two or more kmers that do not exist in a normal database in which the two or more kmers acquired from normal genome information are stored; a position acquisition unit that acquires position information of amino acid sequence information corresponding to each of the one or more kmers determined by the determination unit, using the position information acquired by the mapping unit, which is position information of the read corresponding to the kmer corresponding to the amino acid sequence information; a neoantigen acquisition unit that acquires neoantigen information that is information that associates the position information acquired by the position acquisition unit with each of the one or more amino acid sequence information; and a neoantigen output unit that outputs the neoantigen information.

[0230] FIG. 19 shows the appearance of a computer that executes the program described in this specification to realize the neoantigen prediction device A and the like of the various embodiments described above. The above-mentioned embodiments can be realized by computer hardware and a computer program executed thereon. FIG. 19 is an overview of this computer system 300, and FIG. 20 is a block diagram of the system 300.

[0231] In FIG. 19, a computer system 300 includes a computer 301 including a CD-ROM drive, a keyboard 302, a mouse 303, and a monitor 304.

[0232] 20, computer 301 includes, in addition to CD-ROM drive 3012, MPU 3013, bus 3014 connected to CD-ROM drive 3012 etc., ROM 3015 for storing programs such as a boot-up program, RAM 3016 connected to MPU 3013 for temporarily storing instructions of application programs and providing temporary storage space, and hard disk 3017 for storing application programs, system programs, and data. Although not shown here, computer 301 may further include a network card for providing connection to a LAN.

[0233] A program for causing the computer system 300 to execute the functions of the neoantigen prediction device A of the above-mentioned embodiment may be stored on a CD-ROM 3101, inserted into the CD-ROM drive 3012, and further transferred to the hard disk 3017. Alternatively, the program may be transmitted to the computer 301 via a network (not shown) and stored on the hard disk 3017. The program is loaded into the RAM 3016 when executed. The program may be loaded directly from the CD-ROM 3101 or the network.

[0234] The program does not necessarily include an operating system (OS) or a third-party program that causes the computer 301 to execute the functions of the neoantigen prediction device A of the above-mentioned embodiment. The program only needs to include an instruction portion that calls appropriate functions (modules) in a controlled manner to obtain a desired result. How the computer system 300 operates is well known, and a detailed description will be omitted.

[0235] In addition, in the above program, the steps of transmitting information and receiving information do not include processing performed by hardware, such as processing performed by a modem or interface card in the transmitting step (processing that is performed only by hardware).

[0236] The program may be executed by a single computer or a plurality of computers, that is, centralized processing or distributed processing may be performed.

[0237] Furthermore, in each of the above embodiments, it goes without saying that two or more communication means present in one device may be physically realized by one medium.

[0238] Furthermore, in each of the above embodiments, each process may be realized by centralized processing in a single device, or may be realized by distributed processing in a plurality of devices.

[0239] The present invention is not limited to the above-described embodiment, and various modifications are possible, and it goes without saying that these modifications are also included within the scope of the present invention. [Industrial Applicability]

[0240] As described above, the neoantigen prediction device A of the present invention has the effect of being able to predict neoantigens with high accuracy while obtaining location information of neoantigens with high accuracy, and is useful as a server or standalone device for predicting neoantigens. [Explanation of symbols]

[0241] A Neoantigen Prediction Device 1 Storage area 2. Reception 3 Processing section 4 Output section 11 Reference Genome Sequence Management Department 12 Normal Database Management Department 13 Cipher table management section 14 Range Management Department 21 Target Reception Department 22 Range Reception Section 31 Mapping Section 32 Selection section 33 Decision Section 34 Translation Department 35 Position acquisition part 36 Neoantigen Acquisition Department 41 Neoantigen output section 331 kmer acquisition method 332 Decision-making procedures 333 Counting Methods 334 kmer selection method 351 Score Acquisition Method 352 Neoantigen Selection Methods 353 Location acquisition means

Claims

1. a subject receiving unit that receives one or more reads that are a portion of a first length in subject genome information that includes abnormal genome information; a mapping unit that acquires position information, which is information for identifying the position of each of the one or more reads, by referring to reference genome sequence information that is normal genome information and is a source for acquiring position information of the reads constituting the genome information; a determination unit that obtains two or more kmers that are portions of a second length (first length>=second length) from each of the one or more reads, and determines, from the two or more kmers, one or more kmers that do not exist in a normal database in which the two or more kmers obtained from normal genome information are stored; a position acquisition unit that acquires position information of amino acid sequence information corresponding to one or more kmers determined by the determination unit, the position information being position information of the read corresponding to the kmer corresponding to the amino acid sequence information, using the position information acquired by the mapping unit; a neoantigen acquisition unit that acquires neoantigen information, which is information that associates the position information acquired by the position acquisition unit with each of the one or more pieces of amino acid sequence information; A neoantigen prediction device comprising a neoantigen output unit that outputs the neoantigen information.

2. a range receiving unit that receives range information that identifies a range to be inspected in the target genome information; a selection unit that selects one or more leads corresponding to a position in the range specified by the range information using the position information of the leads acquired by the mapping unit, The determination unit is The neoantigen prediction device according to claim 1 , wherein the selection unit determines one or more kmers that do not exist in the normal database from the one or more reads selected.

3. The determination unit is kmer acquisition means for acquiring one or more kmers from each of the one or more leads; a determination means for determining one or more kmers that are not present in the normal database from among the one or more kmers acquired by the kmer acquisition means; A counting means for acquiring the number of occurrences of each of the one or more kmers in each of the one or more reads; The neoantigen prediction device according to claim 1 or claim 2, further comprising a kmer selection means for selecting one or more kmers whose occurrence counts satisfy a count condition from the one or more kmers determined by the determination means.

4. a translation unit that refers to a DNA genetic code table, which is a correspondence table between character strings constituting kmers and characters constituting amino acid sequences, and obtains amino acid sequence information that specifies an amino acid sequence for each of the one or more kmers determined by the determination unit; The position acquisition unit is The neoantigen prediction device according to any one of claims 1 to 3, wherein the position information of each of the one or more amino acid sequence information acquired by the translation unit is acquired using the position information of the lead corresponding to the kmer corresponding to the amino acid sequence information.

5. The position acquisition unit is a score acquisition means for acquiring a score for each of the one or more pieces of amino acid sequence information acquired by the translation unit, the score being related to whether the amino acid sequence information is a neoantigen; A neoantigen selection means for acquiring one or more pieces of amino acid sequence information whose score satisfies the neoantigen condition; The neoantigen prediction device according to any one of claims 1 to 4, further comprising a position acquisition means for acquiring position information of each of the one or more amino acid sequence information acquired by the neoantigen selection means using position information of the lead corresponding to the kmer corresponding to the amino acid sequence information.

6. A neoantigen prediction method realized by a target receiving unit, a mapping unit, a determination unit, a position acquisition unit, a neoantigen acquisition unit, and a neoantigen output unit, a subject receiving step in which the subject receiving unit receives one or more reads that are a portion of a first length in the subject genome information including abnormal genome information; a target receiving step in which the target receiving unit refers to reference genome sequence information that is normal genome information and is a source of acquiring position information of the leads constituting the genome information, and acquires position information that is information for identifying the position of each of the one or more reads; a subject acceptance step in which the subject acceptance unit acquires two or more kmers that are portions of a second length (first length>=second length) from each of the one or more reads, and determines one or more kmers among the two or more kmers that do not exist in a normal database in which two or more kmers acquired from normal genome information are stored; a position acquisition step in which the position acquisition unit acquires position information of amino acid sequence information corresponding to each of one or more kmers determined by the determination unit, the position information being position information of the read corresponding to the kmer corresponding to the amino acid sequence information, using the position information acquired by the mapping unit; a neoantigen acquisition step in which the neoantigen acquisition unit acquires neoantigen information, which is information that associates the position information acquired by the position acquisition unit with each of the one or more pieces of amino acid sequence information; A neoantigen prediction method comprising: a neoantigen output step in which the neoantigen output unit outputs the neoantigen information.

7. Computer, a subject receiving unit that receives one or more reads that are a portion of a first length in subject genome information that includes abnormal genome information; a mapping unit that acquires position information, which is information for identifying the position of each of the one or more reads, by referring to reference genome sequence information that is normal genome information and is a source for acquiring position information of the reads constituting the genome information; a determination unit that obtains two or more kmers that are portions of a second length (first length>=second length) from each of the one or more reads, and determines, from the two or more kmers, one or more kmers that do not exist in a normal database in which the two or more kmers obtained from normal genome information are stored; a position acquisition unit that acquires position information of amino acid sequence information corresponding to one or more kmers determined by the determination unit, the position information being position information of the read corresponding to the kmer corresponding to the amino acid sequence information, using the position information acquired by the mapping unit; a neoantigen acquisition unit that acquires neoantigen information, which is information that associates the position information acquired by the position acquisition unit with each of the one or more pieces of amino acid sequence information; A program for functioning as a neoantigen output unit that outputs the neoantigen information.

Images