JP7089804B2 - A storage medium that stores a data creation device, a data creation method, and a data creation program. - Google Patents

Description

The present invention relates to a data creation device that creates data that can reproduce the reading frequency by a genome sequencer, a data creation method, and a storage medium that stores a data creation program.

It is expected that the genomic information of living organisms will be utilized for various purposes.

For example, analyzing the genetic constitution of a human or animal based on the genomic information of a human or an animal, predicting the onset of a disease of a human or an animal, or grasping the progress of a disease of a human or an animal. Is expected. It is also expected to optimize soil, water or products based on the genomic information of plants or microorganisms.

In utilizing such genomic information, it is necessary to collect a large amount of genomic information. However, in general, data showing genomic information often has a very large data capacity. For example, the sequence group data for reproducing the human genome, which is the human genome information, reaches a data capacity of several hundred gigabytes.

Therefore, if all the genomic information is stored or transmitted as it is, the storage capacity of the database may be compressed or the communication line may be tight.

Therefore, it is an important issue to reduce the data capacity of genomic information.

In Patent Document 1, the reference genome data and each person's genome data are compared, and only the base information different between the reference genome data and each person's genome data is stored and transmitted. A technique for compressing the data volume to about 0.1% of the genomic data has been proposed.

International Publication No. 2015/146852

However, the technique of Patent Document 1 only reproduces the sequence of base symbols (sequence of ACGT) of the genome data of each person. That is, the technique of Patent Document 1 could not reproduce information other than the base symbol, for example, the frequency of reading the base information by the genome sequencer.

In general, when a genome sequencer reads a target genome information, a single reading does not mean the entire genome information (about 3.1 billion base pairs in the case of a human), but a part of the genome information data (hereinafter referred to as “)”. Read "read" as appropriate). The base sequence contained in the read read by one reading is, for example, about 50 base pairs.

The genome sequencer is configured to read reads repeatedly until the base sequence contained in the read reads can reconstruct all of the genomic information.

Here, the genome sequencer does not always read the read uniformly over the entire genomic information, and reads it frequently in one place and infrequently in another place. As a result, the frequency of reading each base sequence may vary.

The frequency of reading by the genome sequencer is a useful index for determining the site to which molecular modification of the genome and the interaction protein binds and its statistical significance. By analyzing the variation in reading frequency by the genome sequencer, it is possible to obtain information other than the reproduction of the sequence of base symbols.

However, as described above, the technique of Patent Document 1 could not reproduce the reading frequency by the genome sequencer.

Therefore, an object of the present invention is to provide a data creation device for creating data capable of reproducing the reading frequency by a genome sequencer, a data creation method, and a storage medium for storing a data creation program while suppressing the data capacity.

The data creation device of the present invention has a first base sequence storage unit that stores first base sequence data having a length of the first base sequence, and the first base sequence data stored in the first base sequence storage unit. Based on this, for each second base sequence data whose individual length is a second base number shorter than the first base number, the position of the partial sequence in the first base sequence data corresponding to the second base sequence data. The position recognition unit that recognizes the numerical value indicating the value and the numerical value indicating the position of the partial sequence of the first base sequence data corresponding to each second base sequence data are rearranged in ascending or descending order to create a position sequence. A replacement unit, a reference element recognition unit that recognizes a reference element that is at least one element included in the array at the position, and a difference recognition unit that recognizes a difference between adjacent elements included in the array at the position. It is characterized by including a data creation unit that creates data including a reference element recognized by the reference element recognition unit and a difference between the elements recognized by the difference recognition unit.

According to the data creation device having the above configuration, the position recognition unit stores the first base sequence data in the first base sequence storage unit, and the individual lengths of the second base sequence data are shorter than the first base sequence data. For each second base sequence data which is the number of bases, a numerical value indicating the position of the partial sequence in the first base sequence data corresponding to the second base sequence data is recognized.

Then, the rearrangement unit rearranges the numerical values indicating the positions of the partial sequences of the first base sequence data corresponding to each second base sequence data in ascending or descending order, so that the first base sequence data corresponding to each second base sequence data can be sorted. An array of the positions of the partial sequences of the base sequence data is created. Here, since the adjacent elements of the sequence at the position of the partial sequence of the first base sequence data corresponding to each second base sequence data are arranged in ascending order or descending order, the difference tends to be considerably small. In particular, for the base sequence data related to the base sequences frequently read by the genome sequencer, the numerical values indicating their positions are the same or almost the same.

Then, the reference element recognition unit recognizes the reference element which is at least one element included in the array at the position.

Then, the difference recognition unit recognizes the difference between the elements included in the array at the position and adjacent to each other.

Then, the data creation unit creates data including the reference element recognized by the reference element recognition unit and the difference between the elements recognized by the difference recognition unit.

As described above, the difference between the elements related to the base sequence frequently read by the genome sequencer tends to be considerably small, so that the data capacity of the data indicating the difference between the elements can be kept small.

On the other hand, if the reference element included in the created data and the difference between the elements are used, the numerical value indicating the position of the partial sequence of the first base sequence data corresponding to each second base sequence data can be calculated back. can. The numerical value indicating the position of the partial sequence of the first base sequence data corresponding to each of the second base sequence data is such that the base sequence of which part of the target genomic data is read at what frequency. Will be shown.

As described above, according to the data creation apparatus of the present invention, it is possible to create data that can reproduce the reading frequency by the genome sequencer while suppressing the data capacity.

In the data creation device of the present invention, it is preferable that the reference element recognition unit is configured to recognize the element having the smallest value among the elements included in the position array as the reference element.

According to the data creation device having the configuration, the element having the smallest value among the elements included in the position array is recognized as the reference element by the reference element recognition unit. As a result, the data capacity of the data indicating the reference element can be suppressed to a small size, so that the compression rate can be further improved.

In the data creation device of the present invention, the data creation unit uses the first part indicating whether or not the preceding or succeeding data is related data and the data of 14 bits or less as the data indicating the difference between the elements. It is preferable to create variable length data including one or more second parts to be stored.

According to the data creation device having the configuration, the data creation unit uses the first part indicating whether or not the preceding or succeeding data is related data as the data indicating the difference between the elements, and 14 bits or less. Variable length data is created that includes one or more second parts that store the data.

According to the applicants of the present application, the difference between consecutive elements of the array of positions can be represented by almost 14 bits or less. This makes it possible to express the difference between each element of a large amount of data while keeping the data capacity small.

Further, the first part indicating whether the preceding or succeeding data is related data indicates that an appropriate number of second parts are used as data indicating the difference between the elements. Differences that are equal to or greater than the number of bits in the two parts can also be expressed by the variable length data.

In the data creation device having the configuration, the second portion is preferably 6 bits or less.

According to the examination of the applicant, it was found that about 80% of the data can be expressed by 6 bits or less as the difference between the elements.

By expressing the data of the second part with 6 bits or less, it is possible to further reduce the data capacity for a large amount of data. On the other hand, the first part, which indicates whether the preceding or succeeding data is related data, indicates that an appropriate number of second parts are used as the data indicating the difference between the elements. Differences that are equal to or greater than the number of bits in the two parts can also be expressed by the variable length data.

In the data creation device having the configuration, the second portion is preferably 3 bits or less.

According to the examination of the applicant, it was found that about 60% of the data can be expressed in 3 bits or less. By expressing the data of the second part with 3 bits or less, it is possible to further reduce the data capacity for a large amount of data. On the other hand, the first part, which indicates whether the preceding or succeeding data is related data, indicates that an appropriate number of second parts are used as the data indicating the difference between the elements. Differences that are equal to or greater than the number of bits in the two parts can also be expressed by the variable length data.

Overall configuration diagram of the data creation system. The figure which shows an example of the 1st base sequence data. The figure which shows an example of a plurality of 2nd base sequence data read by a genome sequencer. Flowchart of data creation process. The figure which shows an example of the file of SAM format. The figure which shows an example of the data after extraction. The figure which shows an example of the data after sorting. The figure which shows an example of the data after difference recognition. The figure which shows an example of the contents contained in the data created by the data creation process. The figure which shows the specific example of the data created by the data creation process. The figure which shows an example of the format of the data created by the data creation process. The figure which shows the representation of data according to one data format. A graph showing the relationship between the number of bits required to represent a difference and the frequency and content ratio of each bit number.

The data creation system of the embodiment of the present invention will be described with reference to FIGS. 1 to 8.

(Configuration of data creation system)
The configuration of the data creation system will be described with reference to FIG.

The data creation system includes one or more genome sequencers 100, one or more data creation devices 200, and a database 300.

The one or more data generation devices 200 are connected to each of the one or more genomic sequencers 100 via a wired or wireless connection, respectively. The database 300 is connected to each of the data creation devices 200 via a wide area network such as the Internet. The one or more data creation devices 200 may be used by different users.

(Construction of genome sequencer)
For example, the genome sequencer 100 acquires a part of the genomic information from the target living body P, and repeats data showing a partial base sequence included in the genome information (hereinafter, referred to as “second base sequence data”). It is configured to output. The genome sequencer 100 is composed of, for example, the HiSeq system (registered trademark). The second base sequence data is represented by repeating the base symbol (A, C, G or T). The genome sequencer 100 reads the second base sequence data so that the number of base symbols is included in the predetermined setting or the number specified by the user. Hereinafter, the number of base symbols included in the second base sequence data is also appropriately referred to as "length of the second base sequence data". The second base sequence data may include a code other than the base symbol, for example, "?" As a symbol indicating unreadable. The length of the second base sequence data corresponds to an example of the "second base number" of the present invention.

(Configuration of data creation device)
The one or more data creating devices 200 have the following configurations, although they differ in detail for each terminal.

The data creation device 200 includes an arithmetic processing unit 210 and a storage unit 220.

The data creation device 200 may be composed of a computer whose size, shape and weight are designed so that it can be carried by a user, such as a laptop computer, a tablet terminal or a smartphone, and a specific location such as a desktop computer. It may be configured by a computer whose size, shape and weight are designed to be installed in.

The arithmetic processing unit 210 is composed of an arithmetic processing device such as a CPU (Central Processing Unit), a storage device such as a memory, and an I / O (Input / Output) device. A data creation program 223 downloaded from the outside is installed in the storage unit 220. When the data creation program 223 stored in the storage unit 220 is activated, the arithmetic processing unit 210 includes a position recognition unit 211, a sort unit 212, a reference element recognition unit 213, a difference recognition unit 214, and data. It is configured to function as a creation unit 215. The data creation device 200 that stores the data creation program 223 corresponds to an example of the "storage medium" of the present invention.

The arithmetic processing unit 210 is configured to communicate with an external device such as a database 300 via wired communication or wireless communication according to a communication standard suitable for long-distance wireless communication such as WiFi (registered trademark). There is.

The storage unit 220 includes, for example, a storage device such as a ROM (Read Only Memory), a RAM (Random Access Memory), and an HDD (Hard Disk Drive).

The storage unit 220 is configured to store information recognized by the arithmetic processing unit 210, such as arithmetic processing by the arithmetic processing unit 210 and data received by the arithmetic processing unit 210.

Note that one device "recognizes" information means that one device receives the information from another device, and that one device receives information stored in a storage medium connected to the one device. Reading, one device acquiring information based on the signal output from the sensor connected to the one device, one device acquiring from the received information or information stored in a storage medium or sensor. Deriving the information by executing a predetermined arithmetic process (calculation process, search process, etc.) based on the information It means that all arithmetic processing for acquiring the information is executed, such as receiving from, one device reading the information from the internal storage device or the external storage device according to the received signal.

The storage unit 220 includes a first base sequence storage unit 221 and a data storage unit 222.

As shown in FIG. 2A, the first base sequence storage unit 221 stores data indicating the base sequence (hereinafter, referred to as “first base sequence data”). These data can be created based on the data showing each base sequence read from one or more living organisms (however, living organisms having some common items such as "human beings" or "Japanese"). When data showing a base sequence is created from a plurality of living organisms, the first base sequence data is represented by the same base symbol for the bases common to each base sequence read from each living body, and differs among those living bodies. The base is represented by a symbol different from the base symbol such as *. The first base sequence data may be decomposed into a plurality of base sequences such as chr1 and chr2 and stored for each base sequence. Each of these base sequences decomposed into chr1, chr2 and the like is hereinafter appropriately referred to as a "reference sequence". Further, the character string that identifies each of the base sequences of chr1 and chr2 is hereinafter appropriately referred to as "reference sequence name". In this embodiment, the name of the reference sequence is composed of a predetermined character string such as chr and a number. The length of these reference sequences is set longer than the length of the second base sequence data. The total value of the lengths of the reference sequences corresponds to an example of the "first number of bases" of the present invention.

The first base sequence storage unit 221 may store the first base sequence data for each type of living body.

The living body as a sample for creating the first base sequence data is a living body different from the living body P to be processed in the data creation process described later. However, if the types of living organisms are common, most of the base sequences will be the same even if the individuals are different. For example, in the case of human beings, even if the individuals are different, about 99.9% of the base sequences will match.

(Database configuration)
The database 300 is composed of an arithmetic processing unit such as a CPU, a storage device such as a local memory, a ROM, a RAM, and an HDD, and an I / O device. The database 300 is configured to store the data received from the data creation device 200. The database 300 may be configured by one processor or may be configured by a plurality of processors capable of intercommunication.

A part or all of the computers constituting the database 300 may be configured by the computers constituting the data creation device 200. For example, a part or all of the database 300 may be configured by one or more data creation devices 200 as mobile stations.

Further, the database 300 is connected to a public communication network (for example, the Internet) as a network via WiFi or a wired connection, and is configured to communicate with an external device (for example, a data creation device 200).

(Data creation process)
Next, with reference to FIGS. 2 to 8, the flow of the data creation process executed by the data creation device 200 will be described.

The position recognition unit 211 recognizes each second base sequence data of the target living body P based on the data output from the genome sequencer 100 (FIG. 3 / STEP002). The target organism P may be any organism whose genomic information can be read by the genome sequencer 100, for example, a human being, an animal, a plant, or a microorganism. There may be.

The data output from the genome sequencer 100 is, for example, data D1 including repetition of the base symbol, as shown in FIG. 2B.

The data D1 includes a plurality of second base sequence data D11, D12, D13 represented by repetition of base symbols for a predetermined number of bases (for example, 50). The second base sequence data D11, D12, and D13 are separated by, for example, a comma. In addition, each of the second base sequence data D11, D12, and D13 includes auxiliary base symbols D111, D121, and D131 indicating unreadable bases.

The position recognition unit 211 compares each second base sequence data recognized in FIG. 3 / STEP002 with the first base sequence data stored in the first base sequence storage unit 221 to obtain each second base sequence data. Recognizes the numerical value indicating the position of the partial sequence of the first base sequence data in the first base sequence data corresponding to (FIG. 3 / STEP004).

For example, the position recognition unit 211 corresponds to a partial sequence of the first base sequence data (corresponding to the second base sequence data) in which the appearance order of each base symbol included in the second base sequence data has the highest matching ratio. The partial sequence of the first base sequence data) is recognized. Then, the position recognition unit 211 recognizes a numerical value indicating the start position of the partial sequence in the first base sequence data. The position of the partial array may be any position as long as it is a position for specifying the partial array, and is not limited to the start position, and may be, for example, an end position or another position.

Various known methods can be adopted for recognizing such numerical values indicating the positions of partial sequences.

The position recognition unit 211 creates a SAM (Sequence Alignment / Map) format file (FIG. 3 / STEP006). The created file is stored in the storage unit 220.

FIG. 4 is a diagram showing an example of a file created in FIG. 3 / STEP006. The file shown in FIG. 4 includes header data D21 and body data D22.

The body data D22 has the name D221 of the reference sequence, the start position D222 of the partial sequence in the reference sequence of the first base sequence data corresponding to the second base sequence data, and the second base sequence for each of the second base sequence data. The start position D223 of the partial sequence in the case of the pair end in the first base sequence data corresponding to the data and the base sequence D224 of the second base sequence data are included. The name of the reference sequence and the start position of the partial sequence in the reference sequence correspond to an example of the "position of the partial sequence of the first base sequence data" of the present invention.

The sorting unit 212 extracts a numerical value indicating the start position of the partial sequence in the first base sequence data corresponding to each second base sequence data for each reference sequence (FIG. 3 / STEP008).

The sorting unit 212 creates, for example, the post-position extraction data D3 shown in FIG. 5A by the process of FIG. 3 / STEP008. The position-extracted data D3 includes the length D31 of the base sequence of each second base sequence data, the name D32 of each reference sequence, the number D33 of the second base sequence data associated with each reference sequence, and each second. It includes the start position D34 of the partial sequence in the first base sequence data corresponding to the two base sequence data. The base sequence length D31 of each second base sequence data may be recognized from the length of each second base sequence data. When the length of the second base sequence data is predetermined, the length D31 of the base sequence of each second base sequence data may be omitted.

In the position-extracted data D3 shown in FIG. 5A, the fifth and subsequent rows are the start positions of the partial sequences in the first base sequence data corresponding to the second base sequence data.

In the position-extracted data D3 shown in FIG. 5A, the data corresponding to each of the names D32 of the reference sequence in the second row are stored in the third and subsequent rows separated by commas.

For example, the first "719786" in the third row indicates the number of second base sequence data associated with the reference sequence "chr1".

Further, the second "380912" in the third row indicates the number of the second base sequence data associated with the reference sequence "chr2".

Further, the first "177644860" in the fourth row is the start of a partial sequence in the first base sequence data corresponding to a certain second base sequence data among the second base sequence data associated with the reference sequence "chr1". It is a numerical value indicating the position.

Further, the first "177644896" in the fifth row is a partial sequence of the first base sequence data corresponding to another second base sequence data among the second base sequence data associated with the reference sequence "chr1". It is a numerical value indicating the start position.

If there is no corresponding start position, it will be blank.

The rearrangement unit 212 replaces the start position of the partial sequence in the first base sequence data corresponding to each second base sequence data for each associated reference sequence with the first base corresponding to the second base sequence data. A numerical value indicating the start position of the partial array in the case of a pair end in the array data may be extracted.

The sorting unit 212 sorts the numerical values indicating the start positions in ascending order for each reference array (FIG. 3 / STEP010).

After the processing of FIG. 3 / STEP010, the sorting unit 212 creates the post-sorting data D4 as shown in FIG. 5B. The rearranged data D4 includes the length D41 of the base sequence of each second base sequence data, the name D42 of each reference sequence, the number D43 of the second base sequence data associated with each reference sequence, and each second. The start positions D44, D45, and D46 of the partial sequence in the first base sequence data corresponding to the two base sequence data are included. The start positions D44, D45, and D46 of the partial sequence in the first base sequence data corresponding to each second base sequence data are rearranged in ascending order. Therefore, the data D44 in the top row (fourth row in the sorted data D4) is the smallest element (start position) in each reference sequence.

The reference element recognition unit 213 recognizes one or a plurality of reference elements for each reference sequence (FIG. 3 / STEP012). The reference element is, for example, the smallest element in each reference sequence. Any element other than the smallest element may be recognized as a reference element. Further, a plurality of elements may be recognized as reference elements in the position reference array.

The difference recognition unit 214 recognizes an array of differences between adjacent elements for elements other than the reference element (FIG. 3 / STEP014). The difference recognition unit 214 creates, for example, the difference recognition post-data D5 shown in FIG. 5C after the processing of FIG. 3 / STEP014.

The difference recognition data D5 includes the length D51 of the base sequence of each second base sequence data, the name D52 of each reference sequence, the number D53 of the second base sequence data associated with each reference sequence, and each. The reference element D54 of the reference sequence and the difference data D55 and D56 are included.

For example, in the rearranged data D4 shown in FIG. 5B, the start positions of the partial sequences included in the reference sequence chr1 are 9997, 9998, 9998, ... In ascending order.

The fourth row (row represented by reference numeral D54) of the difference recognition data D5 shown in FIG. 5C contains the reference element 9997 recognized in FIG. 3 / STEP012 in the reference sequence chr1.

Further, the element corresponding to the reference array chr1 in the fifth row (row represented by reference numeral D55) of the difference recognition data D5 shown in FIG. 5C is the fifth row (row indicated by reference numeral D45) in FIG. 5B. It contains 1 which is a difference between the element 9998 and the element 9997 before it (the fourth line (the line indicated by the reference numeral D44)).

Further, the element corresponding to the reference sequence chr1 in the sixth row (row indicated by reference numeral D55) of the difference recognition data D5 shown in FIG. 5C is the element corresponding to the sixth row (row indicated by reference numeral D46) in FIG. 5B. It contains 0, which is the difference between the element 9998 and the element 9998 before it (the fifth line (the line indicated by the reference numeral D45)).

Further, for example, in the rearranged data D4 shown in FIG. 5B, the start positions of the partial sequences included in the reference sequence chr2 are 10237, 10286, 10330, ... In ascending order.

The fourth row (row represented by reference numeral D54) of the difference recognition data D5 shown in FIG. 5C contains the reference element 10237 recognized in FIG. 3 / STEP012 in the reference sequence chr2.

Further, the element corresponding to the reference array chr2 in the fifth row (row indicated by reference numeral D55) of the difference recognition data D5 shown in FIG. 5C is the fifth row (row indicated by reference numeral D45) in FIG. 5B. 49 is included, which is the difference between the element 10286 and the element 10237 before it (the fourth line (the line indicated by the reference numeral D44)).

Further, the element corresponding to the reference array chr2 in the sixth row (row represented by reference numeral D55) of the difference recognition data D5 shown in FIG. 5C is the sixth row (row indicated by reference numeral D46) in FIG. 5B. 44 is included, which is the difference between the element 10330 and the element 10286 before it (the fifth line (the line indicated by the reference numeral D45)).

The data creation unit 215 creates data including the difference between the reference element recognized in FIG. 3 / STEP012 and the adjacent element recognized in FIG. 3 / STEP014 (FIG. 3 / STEP016).

For example, the data creation unit 215 creates the data D61 as shown in FIG. 6A for each reference sequence in FIG. 3 / STEP016 based on the difference recognition data D5 shown in FIG. 5C. The data D61 includes the length D61 of the base sequence of the second base sequence data, the number D62 included in the name of the reference sequence, the number D63 of the second base sequence data associated with the reference sequence, and the reference element. This is data including D64 and an array of differences D65 and D66.

The data created in FIG. 3 / STEP016 is in a format including the first portion D1 and the second portion D2, as shown in FIG. 7A, at least for the data portion showing the difference.

The second portion D2 may have any number of bits, but is preferably 14 bits or less, more preferably 6 bits or less, and further preferably 3 bits or less.

The first part D1 is a part indicating whether or not the preceding or succeeding data is related data. The second part D2 is a part showing the contents of the target data such as the difference. The first portion D1 may be composed of, for example, one bit.

When the first part D1 is composed of 1 bit, for example, when the first part is 0, it means that the data of the following predetermined length is not related, and when the first part is 1, the succeeding predetermined It may mean that the data of the length of is related, but any rule may be used as long as the range to be read can be specified by the first part.

For example, the data shown in FIG. 7B shows an example in which the first part is composed of 1 bit and the second part is composed of 3 bits. The data shown in FIG. 7B means that when the first part is 0, the subsequent data of a predetermined length is not related, and when the first part is 1, the subsequent data of a predetermined length is related. Means to do.

When the second part is 3 bits, it is not necessary to use the following data because the decimal numbers 1 to 7 can be sufficiently expressed by 3 bits. Therefore, for decimal numbers 1 and 3, the first part is 0 as shown in FIG. 7B. Further, for decimal numbers 1 and 3, as shown in FIG. 7B, the second part is 001 and 011, respectively.

On the other hand, the decimal numbers 8 to 31 cannot be sufficiently expressed by 3 bits. Therefore, for these data, as shown in FIG. 7B, the first part of the first data is 1. However, since 6 bits can sufficiently represent decimal numbers 8 to 31, the first part of the next data is 0 for these data, as shown in FIG. 7B. For these, the contents of the target data such as differences are shown by the entire related second part. For example, in the case of 8, as shown in FIG. 7B, the binary number 8 is represented by 001000, which is the sum of 001 of the first second part and 000 of the next second part.

The size of the second part can be optimized by analyzing the size of the target data.

FIG. 6B is a diagram showing an example when the data shown in FIG. 6A is created in a data format including such a first portion and a second portion.

In FIG. 6B, the number included in the name of the reference sequence, the total number of the second base sequence data associated with the reference sequence, the reference element, and each difference are represented in the above-mentioned data format. .. The data created in FIG. 3 / STEP016 contains the data shown in FIG. 6B repeatedly as many as the number of reference sequences. The total number of the second base sequence data associated with the reference sequence is used to indicate the delimiter for each reference sequence.

The data creation unit 215 stores the created data in the data storage unit 222 in a binary format and transmits the created data to the database 300. The database 300 stores the received data together with the information that can identify the data creation device 200 or the target living body P (for example, a user ID). The data creation unit 215 may send the data excluding the reference sequence and the reference element to the database 300. By doing so, all the data cannot be restored from the data stored in the database 300, so that personal information can be protected.

This completes the data creation process.

(Data restoration)
A method of restoring data from the data created in FIG. 3 / STEP016 will be described. The following processing can be performed by a general computer having access to the first base sequence data.

First, in the first step, the computer reads the data created in FIG. 3 / STEP016 from the beginning, the length of the base sequence of each second base sequence data, the number included in the name of one reference sequence, and the said. Recognize the total number of second base sequence data associated with the reference sequence.

Next, in the second step, the computer recognizes the reference element.

In the third step, the computer can recognize the start position of the partial sequence of the first base sequence data corresponding to the reference element from the number included in the name of one reference sequence and the reference element. The computer can recognize the partial sequence of the first base sequence data corresponding to the reference element based on the start position of the partial sequence and the length of the base sequence of each second base sequence data. Further, the computer subtracts 1 from the total number of the second base sequence data associated with the reference sequence.

In the fourth step, the computer reads the next difference of the reference element. The computer recognizes the value of the second element by adding the difference to the reference element. Based on this value, the computer can recognize the start position of the partial sequence of the first base sequence data corresponding to the second element. The computer can recognize the partial sequence of the first base sequence data corresponding to the second element based on the start position of the partial sequence and the length of the base sequence of each second base sequence data. Further, the computer subtracts 1 from the total number of the second base sequence data associated with the reference sequence.

In the fifth step, the computer reads the next difference. The computer recognizes the value of the third element by adding the difference to the value of the second element. Based on this value, the computer can recognize the start position of the partial sequence of the first base sequence data corresponding to the third element. The computer can recognize the partial sequence of the first base sequence data corresponding to the third element based on the start position of the partial sequence and the length of the base sequence of each second base sequence data. Further, the computer subtracts 1 from the total number of the second base sequence data associated with the reference sequence.

The computer repeats the fifth step until the total number of the second base sequence data associated with the reference sequence becomes zero. When the total number of the second base sequence data becomes zero, the computer repeatedly executes the first step to the fifth step until the reading of the data is completed.

By doing so, the computer can recognize a group of partial sequences of the first base sequence data corresponding to each second base sequence data. The group of partial sequences of the first base sequence data corresponding to each of the second base sequence data does not completely match each second base sequence data, but it is necessary to analyze the reading frequency by the genome sequencer of the living body P. Is useful enough.

(Action and effect of this embodiment)
According to the data creation device 200 having the configuration, each length is shorter than the number of first bases based on the first base sequence data stored in the first base sequence storage unit 221 by the position recognition unit 211. For each second base sequence data having 2 bases, the positions D221 and D222 of the partial sequence in the first base sequence data corresponding to the second base sequence data D224 are recognized (FIG. 3 / STEP004, FIG. 3 /. STEP006).

Then, by rearranging the positions of the partial sequences of the first base sequence data corresponding to each second base sequence data in ascending or descending order by the rearrangement unit 212 (FIG. 3 / STEP010), each second base sequence data can be obtained. A sequence of the positions of the partial sequences of the corresponding first base sequence data (from the fourth row in FIG. 5B) is created (see FIG. 5B). Here, since the adjacent elements of the sequence at the position of the partial sequence of the first base sequence data corresponding to each second base sequence data are located close to each other, the difference tends to be considerably small. In particular, for the base sequence data related to the base sequence read frequently, their positions are the same or almost the same.

Then, the reference element recognition unit 213 recognizes the reference element which is at least one position included in the array of the positions (FIG. 3 / STEP012).

Then, the difference recognition unit 214 recognizes the array of differences between adjacent elements of the array of positions (FIG. 3 / STEP014).

Then, the data creation unit 215 creates data D6 including the reference element recognized by the reference element recognition unit 213 and the difference between the elements recognized by the difference recognition unit 214 (FIG. 3 / STEP016).

Since the difference between the elements tends to be considerably small for the portion read frequently as described above, the data capacity of the data indicating the difference between the elements can be kept small.

For example, according to the experiments conducted by the present inventors, the size of the data created in FIG. 3 / STEP016 was about 0.33% of the size of the SAM file created in FIG.3 / STEP006. Further, the size of the data created in FIG. 3 / STEP016 was about 4.97% even when compared with the size of the file in FIG. 5A from which the numerical value indicating the start position was extracted.

On the other hand, if the reference element included in the created data and the difference between the elements are used, the position of the partial sequence of the first base sequence data corresponding to each second base sequence data can be calculated back. The numerical value indicating the position of the partial sequence of the first base sequence data corresponding to each of the second base sequence data is such that the base sequence of which part of the target genomic data is read at what frequency. Will be shown.

As described above, according to the data creation device 200 of the present invention, it is possible to create data that can reproduce the reading frequency by the genome sequencer 100 while suppressing the data capacity.

Further, according to the data creation device 200 having the above configuration, the reference element recognition unit 213 recognizes the element having the smallest value among the elements included in the position array as the reference element (FIG. 3 / STEP012). As a result, the data capacity of the data indicating the reference element can be suppressed to a small size, so that the compression rate can be further improved.

According to the data creation device 200 having the configuration, the data creation unit 215 uses the first portion D61 indicating whether or not the preceding or succeeding data is related data as the data indicating the difference between the elements, and 14 bits or less. Variable length data D6 (see FIG. 7A) including one or more of the second portion D62 for storing the data of the above is created (FIG. 3 / STEP016).

According to the applicants of the present application, each difference between consecutive elements of the array of positions can be represented by almost 14 bits or less.

For example, FIG. 8 is a graph created based on data acquired from a certain living body (human), and is a graph showing how many bits the difference can be represented. The horizontal axis of the graph of FIG. 8 is an axis indicating how many bits the difference can be represented. The left axis of FIG. 8 is the frequency of appearance of each bit. The right axis of FIG. 8 is the cumulative ratio of the appearance frequency of each bit. As shown in FIG. 8, the cumulative ratio of the appearance frequency of each bit is 14 bits, which is almost 100%. Therefore, the second portion is preferably 14 bits or less.

This makes it possible to express the difference between each element of a large amount of data while keeping the data capacity small.

Further, the first part indicating whether the preceding or succeeding data is related data indicates that an appropriate number of second parts are used as data indicating the difference between the elements. As shown in FIG. 7B, a difference having a number of bits or more in two parts can also be expressed by the variable length data.

Further, as shown in FIG. 8, the cumulative ratio of the appearance frequency of each bit is about 80% for 6 bits. Therefore, the second part may be 6 bits or less.

By expressing the data of the second part with 6 bits or less, it is possible to further reduce the data capacity for a large amount of data. On the other hand, the first part, which indicates whether the preceding or succeeding data is related data, indicates that an appropriate number of second parts are used as the data indicating the difference between the elements. Differences that are equal to or greater than the number of bits in the two parts can also be expressed by the variable length data.

Further, as shown in FIG. 8, the cumulative ratio of the appearance frequency of each bit is approximately 60% for 3 bits. Therefore, the second part may be 3 bits or less.

According to the examination of the applicant, it was found that about 60% of the data can be expressed in 3 bits or less. By expressing the data of the second part with 3 bits or less, it is possible to further reduce the data capacity for a large amount of data. On the other hand, the first part, which indicates whether the preceding or succeeding data is related data, indicates that an appropriate number of second parts are used as the data indicating the difference between the elements. Differences that are equal to or greater than the number of bits in the two parts can also be expressed by the variable length data.

(Deformation mode)
In the above-described embodiment, the first base sequence data is decomposed into a plurality of reference sequences, but the present invention is not limited to this, and the first base sequence data may be represented by a single sequence.

The first part may be 2 bits. In this data format, for example, when the first part is 00, it indicates that the second part is 2 bits, and when the first part is 01, it indicates that the second part is 6 bits. When the first part is 10, the second part is 10 bits, and when the first part is 11, the second part is 10 bits and the subsequent data is related data. May be shown.

Further, the length of the second part may be variable depending on the number of related data. For example, if the number of related data is 1, the second part is 1 bit, and if the number of related data is 2 or more, the second part is 3 bits each. May be good.

The data creation unit 215 may create the data in FIG. 3 / STEP016 according to such a data format.

100 ... Genome sequencer, 200 ... Data creation device, 210 ... Arithmetic processing unit, 211 ... Position recognition unit, 212 ... Sorting unit, 213 ... Reference element recognition unit, 214 ... Difference recognition unit, 215 ... Data creation unit, 220 ... Storage unit, 221 ... First base sequence storage unit, 222 ... Data storage unit, 300 ... Database.

Claims (7)

A first base sequence storage unit for storing first base sequence data having a length of the first base, and a first base sequence storage unit.
Based on the first base sequence data stored in the first base sequence storage unit, each second base sequence data having an individual length shorter than the first base number is the second base sequence data. A position recognition unit that recognizes a numerical value indicating the position of a partial sequence in the first base sequence data corresponding to the base sequence data,
A rearrangement unit that creates a sequence of positions by rearranging the numerical values indicating the positions of the partial sequences of the first base sequence data corresponding to each second base sequence data in ascending or descending order.
A reference element recognition unit that recognizes a reference element that is at least one element included in the array at the position, and a reference element recognition unit.
A difference recognition unit that recognizes the difference between the element and the element immediately preceding the element in the order of arrangement as the difference between adjacent elements for each element after the second in the arrangement order in the arrangement of the positions.
A data creation device including a data creation unit that creates data including a reference element recognized by the reference element recognition unit and a difference between the elements recognized by the difference recognition unit. In the data creation device according to claim 1,
The reference element recognition unit is a data creation device characterized in that it is configured to recognize the element having the smallest value among the elements included in the position array as a reference element. In the data creating apparatus according to claim 1 or 2.
The data creation unit includes, as data indicating the difference between the elements, a first portion indicating whether or not the preceding or succeeding data is related data and a second portion storing data of 14 bits or less. Alternatively, a data creation device characterized in that variable length data including a plurality of data is created. In the data creation device according to claim 3,
The second part is a data creation device characterized by having 6 bits or less. In the data creation apparatus according to claim 4,
The second part is a data creation device characterized by having 3 bits or less. It is a method executed by a computer having a first base sequence storage unit for storing first base sequence data having a length of the first base.
Based on the first base sequence data stored in the first base sequence storage unit, each second base sequence data having an individual length shorter than the first base number is the second base sequence data. A step of recognizing a numerical value indicating the position of a partial sequence in the first base sequence data corresponding to the base sequence data, and
A step of creating a position sequence by rearranging the numerical values indicating the positions of the partial sequences of the first base sequence data corresponding to each second base sequence data in ascending or descending order.
A step of recognizing a reference element, which is at least one element contained in the array at the position,
A step of recognizing the difference between the element and the element one before the element in the order of arrangement as the difference between adjacent elements for each element after the second in the arrangement order in the arrangement of the positions.
A data creation method comprising a step of creating data including the reference element and a difference between the elements. A computer provided with a first base sequence storage unit for storing first base sequence data having a length of the first base.
Based on the first base sequence data stored in the first base sequence storage unit, each second base sequence data having an individual length shorter than the first base number is the second base sequence data. A step of recognizing a numerical value indicating the position of a partial sequence in the first base sequence data corresponding to the base sequence data, and
A step of creating a position sequence by rearranging the numerical values indicating the positions of the partial sequences of the first base sequence data corresponding to each second base sequence data in ascending or descending order.
A step of recognizing a reference element, which is at least one element contained in the array at the position,
A step of recognizing the difference between the element and the element one before the element in the order of arrangement as the difference between adjacent elements for each element after the second in the arrangement order in the arrangement of the positions.
A storage medium containing a data creation program, which comprises executing a step of creating data including the reference element and a difference between the elements.

Images