KR102111731B1 - Method and apparatus for analyzing nucleic acid sequence - Google Patents

Abstract

A method and apparatus for analyzing a nucleic acid sequence acquires information of a first query sequence in which a bulge exists and information of a target sequence of a potential off-target site, and the bulge In order to assume a non-existent sequence, the first query sequence is transformed into a second query sequence in which the base of the bulge is inserted between the bases at both ends of the bulge, and the base positions of the converted second query sequence and the target sequence are matched. The target sequence is obtained by acquiring a second mismatch result in a state where the first mismatch result and base sites are inconsistent, and comparing the obtained first and second mismatch results with a predefined mismatch threshold number. It is determined whether to select as a candidate for determining whether it is an on-target site or an off-target site.

Description

Method and apparatus for analyzing nucleic acid sequence

It relates to a method and apparatus for analyzing nucleic acid sequences, and specifically to a method and apparatus for analyzing mismatches between nucleic acid sequences in which bulges are present.

Methods for regulating DNA or gene expression using short RNA (short guiding RNA or small interfering RNA) that are sequence-specifically bound to a specific DNA position or molecular material have been widely used in the field of genetic engineering. Unexpected binding of off-target sites between nucleic acid sequences can lead to enormous failure costs and side effects in the field of diagnosis / treatment using genetic engineering. If the off-target sites can be excluded in advance by predicting whether to bind to the off-target site effectively, it can make an important contribution to maximizing functions and minimizing side effects in the genome analysis technology.

Disclosed is a method and apparatus for analyzing nucleic acid sequences. The technical problem to be solved is not limited to the technical problems as described above, and other technical problems may exist.

In the genome, tens of millions of sequence candidates (ie, target sequences) that are mismatch calculation targets satisfying a specific condition may exist. Therefore, calculating the mismatch between a sequence of on-target sites (ie, a query sequence) targeting binding and tens of millions of sequence candidates (target sequences) not only takes a long time to process. , The processing speed is also slow.

According to one aspect, a method of analyzing a nucleic acid sequence includes obtaining information of a first query sequence in which a bulge exists and information of a target sequence of a potential off-target site; Converting the first query sequence into a second query sequence in which the base of the bulge is inserted between bases at both ends of the bulge in order to assume the first query sequence as a sequence in which the bulge does not exist; Obtaining a first mismatch result in which the base positions of the converted second query sequence and the target sequence are matched and a second mismatch result in which the base positions are mismatched; By comparing the obtained first and second mismatch results with a predefined number of mismatch thresholds, the target sequence is selected as a candidate for determining whether it is an on-target site or an off-target site. And determining whether or not.

Also, the bulge may be one in the first query sequence.

In addition, the first mismatch result is a mismatch base between bases located on the first side of the base of the bulge included in the second query sequence and bases of the target sequence, with the base positions coinciding. Corresponds to the number.

Also, a state in which the base positions are mismatched corresponds to a state in which the second query sequence is misaligned by one base to the first side with respect to the target sequence, and the second mismatch result is in the second query sequence. Corresponds to the number of mismatch bases between bases located on the second side of the base of the bulge and bases of the target sequence.

Further, when the first side is the 5 'direction of the target sequence, the second side is the 3' direction of the target sequence, and when the first side is the 3 'direction, the second side is the 5' direction Can be

In addition, the nucleic acid sequence comprises a plurality of potential off-target sites, and the method is based on the first and second mismatch results for the target sequence of each of the plurality of potential off-target sites, And selecting the candidate for determining whether it is the on-target site or the off-target site among target sequences of a plurality of potential off-target sites.

In addition, the screening step selects the candidate by pruning target sequences having a number of mismatch bases exceeding the mismatch threshold number among the target sequences of the plurality of potential off-target sites, and the mismatch The number of bases corresponds to the sum of the first mismatch result and the second mismatch result.

In addition, the first query sequence corresponds to the nucleic acid sequence of sgRNA for the CRISPR Cas9 system.

Also, the number of bases included in the target sequence is the same as the number of bases included in the first query sequence.

According to another aspect, an apparatus for analyzing a nucleic acid sequence includes a memory for storing information of a first query sequence in which a bulge is present and information of a target sequence at a potential off-target site; And base information of the bulge between bases of both ends of the bulge in order to obtain information of the first query sequence and information of the target sequence from the memory, and to assume the first query sequence as a sequence in which the bulge does not exist. The first query sequence is converted into the second query sequence inserted in the first mismatch result in which the base positions of the converted second query sequence and the target sequence are matched, and the base positions are inconsistent. Whether the target sequence is an on-target site or off by acquiring a second mismatch result at and comparing the obtained first and second mismatch results with a predefined number of mismatch thresholds It includes a processor that determines whether to select as a candidate for determining whether it is a target site.

Also, the bulge may be one in the first query sequence.

In addition, the first mismatch result is a mismatch base between bases located on the first side of the base of the bulge included in the second query sequence and bases of the target sequence, with the base positions coinciding. Corresponds to the number.

Also, a state in which the base positions are mismatched corresponds to a state in which the second query sequence is misaligned by one base to the first side with respect to the target sequence, and the second mismatch result is in the second query sequence. Corresponds to the number of mismatch bases between bases located on the second side of the base of the bulge and bases of the target sequence.

Further, when the first side is the 5 'direction of the target sequence, the second side is the 3' direction of the target sequence, and when the first side is the 3 'direction, the second side is the 5' direction Can be

In addition, the nucleic acid sequence includes a plurality of potential off-target sites, and the processor is based on the first and second mismatch results for the target sequence of each of the plurality of potential off-target sites, The candidate for determining whether it is the on-target site or the off-target site is selected from target sequences of a plurality of potential off-target sites.

In addition, the processor selects the candidate by pruning target sequences having a number of mismatch bases exceeding the mismatch threshold number among the target sequences of the plurality of potential off-target sites, and the number of mismatch bases Is the sum of the first mismatch result and the second mismatch result.

In addition, the first query sequence corresponds to the nucleic acid sequence of sgRNA for the CRISPR Cas9 system.

Also, the number of bases included in the target sequence is the same as the number of bases included in the first query sequence.

According to the above, it is possible to improve the mismatch processing speed for measuring the degree of binding in the field of genetic engineering using the property that one end of two nucleic acid sequences (binding domain position of a specific protein) is fixed to each other. have. In particular, by improving the speed of mismatch calculation in which a bulge exists between two nucleic acid sequences, mismatch calculation between large nucleic acid sequences such as genomic big data can be enabled in a short time.

1 is a block diagram showing a hardware configuration of a sequence analysis device according to an embodiment.
2 is a diagram for explaining an off-target site and an on-target site according to an embodiment.
3 is a view for explaining the binding of sgRNA with bulges present according to an embodiment.
4A and 4B are diagrams for explaining the running time of each mismatch number determination when there is no bulge and when there is bulge in the query sequence of sgRNA according to one embodiment.
5 is a diagram for explaining a method of calculating a mismatch between a sgRNA sequence and a DNA sequence in which a bulge is present, according to an embodiment.
6 and 7 are diagrams for explaining a first pre-processing process and a second pre-processing process for analyzing actual mismatch results between nucleic acid sequences according to an embodiment.
FIG. 8 is a diagram illustrating that a candidate group for determining an on-target site (or for determining an off-target site) may be reduced by preprocessing performed in FIGS. 6 and 7 according to an embodiment. .
9 is a flowchart of a method for analyzing a nucleic acid sequence according to an embodiment.

The terminology used in the embodiments has been selected for general terms that are currently widely used while considering functions in the present invention, but this may vary according to the intention or precedent of a person skilled in the art or the appearance of new technologies. In addition, in certain cases, some terms are arbitrarily selected by the applicant, and in this case, their meanings will be described in detail in the description of the applicable invention. Therefore, the terms used in the present invention should be defined based on the meanings of the terms and the contents of the present invention, not simply the names of the terms.

When a certain part of the specification “includes” a certain component, this means that other components may be further included instead of excluding the other component, unless specifically stated to the contrary. Also, “… Wealth ”,“… The term “module” means a unit that processes at least one function or operation, which may be implemented in hardware or software, or a combination of hardware and software.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art to which the present invention pertains may easily practice. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

1 is a block diagram showing a hardware configuration of a sequence analysis device according to an embodiment.

Referring to FIG. 1, the sequence analysis device 10 includes a processor 110 and a memory 120. The processor 110 may include a mismatch determination unit 112 and a candidate sequence selection unit 114.

In the sequence analysis apparatus 10 shown in FIG. 1, only components related to the present embodiment are illustrated. Therefore, it can be understood by those skilled in the art related to the present embodiment that other general-purpose components other than those shown in FIG. 1 may be further included in the sequence analysis device 10.

The sequence analysis apparatus 10 may correspond to a genome analysis apparatus that predicts actual off-target sites or target sequence candidates of actual on-target sites by analyzing mismatches between nucleic acid sequences in which bulges exist.

Mismatch processing between nucleic acid sequences introduced in conventional genetic engineering techniques is mostly performed without considering the bulge generated in the nucleic acid sequence. In addition, even in the case of performing mismatch processing in consideration of bulge, an inefficient Edit Distance algorithm method is used for all candidate nucleic acid sequences to be compared, and processing is performed to target big data such as genomes. Time or processing was forced to be slow. In other words, since the candidate group for performing mismatch calculation on the genome is too large, the application of the Edit Distance algorithm to the candidate group of all candidate nucleic acid sequences can be very inefficient.

On the other hand, it is known that it is extremely rare that two or more bulges exist in genetic engineering using binding between short nucleic acid sequences. Accordingly, the sequence analysis apparatus 10 of the present embodiment does not have any bulges or performs mismatch processing in consideration of only one bulge, so that the target sequences of candidate sequences of a full-scale off-target site (or on-target site) are missed It is possible to significantly reduce the number of target sequences in the candidate group before calculating the match. The off-target site and the on-target site are as follows.

2 is a diagram for explaining an off-target site and an on-target site according to an embodiment.

Referring to FIG. 2, sgRNA (small guidingRNA, short guidingRNA, or single guideRNA) 210 may be bound to DNA 220 (specifically, one strand of double-stranded DNA). In this case, the sgRNA 210 may be combined with a specific protein and other types of proteins. For example, the specific protein can be Cas9, in which case sgRNA 210 can be for the CRISPR Cas9 system. However, the production of sgRNA 210 according to this embodiment is not limited to this type. The size of the sgRNA 210 may be, for example, 20 nucleotides (nts) in length, but this embodiment is not limited thereto.

As shown in Figure 2, the portion of the DNA 220 to be bound to the sgRNA 210 corresponds to the on-target site 235, and unintentionally the DNA 220 to which the sgRNA 210 is bound The portion is a portion where one or more base mismatches are generated, and corresponds to the off-target site 233. The NGG sequence corresponds to a PAM (protospacer adjacent motif).

The sequence analysis device according to the present embodiment (10 in FIG. 1) relates to a technique for reducing the pool of candidate groups of the off-target site 233 on which the sgRNA 210 can be bound on the DNA 220. . The candidate group of the selected off-target site 233 may eventually be the candidate group of the final mismatch calculation.

Referring back to FIG. 1, the memory 120 stores information of a first query sequence in which bulge exists and information of a target sequence of a potential off-target site. Such information is received from an external sequencing device (for example, image analysis devices such as a High Content Cell Imaging device, a High Content Screening device, or a High Throughput Screening device), or an external nucleic acid sequence storage database (DB) It may be received from, or may be nucleic acid sequence information obtained using the sequence analysis device 10 itself.

The memory 120 includes various types of random access memory (RAM), read-only memory (ROM), and electrically erasable programmable read-only memory (EEPROM) for storing various data processed in the sequence analysis device 10 Memory.

The processor 110 is hardware that controls the overall operation of the sequence analysis device 10. The processor 110 may correspond to an integrated circuit or processing module implemented as a combination of a processing unit and a memory unit such as at least one processor, microprocessor, microcontroller, and the like, for example, a central processing unit (CPU), GPU ( graphic processing unit).

The mismatch determination unit 112 of the processor 110 obtains information of a first query sequence in which bulge exists and information of a target sequence of a potential off-target site from the memory 120. Here, it is assumed that there is one bulge in the first query sequence, but this embodiment may not be limited thereto. First, the bulge will be described with reference to FIG. 3.

3 is a view for explaining the binding of sgRNA with bulges present according to an embodiment. Referring to FIG. 3, when sgRNA 310 is bound on DNA 220, bulge 312 for one base may be generated. When one bulge 312 is generated in 20 nts of sgRNA 310, it may be bound to some sequence in 19 nts of DNA 220 that is complementary to sgRNA 310. This phenomenon is a mismatch due to bulge.

The first query sequence having a bulge may correspond to the nucleic acid sequence of sgRNA for the CRISPR Cas9 system, but is not limited thereto, and other types of systems may correspond to the nucleic acid sequence of sgRNA or other target gRNA (guide RNA).

The target sequence on DNA 220 may be a portion of the genome or a sequence defined by the user.

Referring back to FIG. 1, the mismatch determination unit 112, based on the obtained information, in order to assume the first query sequence as the sequence in which the bulge does not exist, the base of the bulge between the bases of the bulge ends. The first query sequence is converted into the inserted second query sequence. Then, the mismatch determination unit 112 obtains the first mismatch result in which the base positions of the converted second query sequence and the target sequence are matched, and the second mismatch result in which the base positions are mismatched. do.

The first mismatch result corresponds to the number of mismatch bases between bases located on the first side of the base of the bulge included in the second query sequence and bases of the target sequence, with the base positions coincident.

The second mismatch result corresponds to the number of mismatch bases between the bases located on the second side of the base of the bulge and the bases of the target sequence in the second query sequence. At this time, the state in which the base positions are inconsistent corresponds to a state in which the second query sequence with respect to the target sequence is misaligned by one base to the first side.

The directions of the first side and the second side mean opposite directions in the nucleic acid sequence, and specifically, when the first side is the 5 'direction of the target sequence, the second side is the 3' direction of the target sequence, and the first side is 3 In the 'direction, the second side may be in the 5' direction.

The candidate sequence selector 114 compares the obtained first and second mismatch results with a predefined number of mismatch thresholds, thereby determining whether the target sequence is an on-target site or an off-target site. It is determined whether to select as a candidate for discrimination.

The sequence analysis device 10 is a device for analyzing target sequences of a plurality of potential off-target sites, as described above, before mismatch calculation of target sequences of candidate groups of a full-fledged off-target site (or on-target site) Pruning of target sequences considered unnecessary within the candidate group may be performed.

That is, the candidate sequence selector 114 comes from among target sequences of a plurality of potential off-target sites based on first and second mismatch results for the target sequence of each of the plurality of potential off-target sites. -Candidates are selected to determine whether they are target sites or off-target sites. At this time, the candidate sequence selector 114 may select a candidate by pruning target sequences having a number of mismatch bases exceeding a mismatch threshold number among target sequences of a plurality of potential off-target sites. Here, the number of mismatch bases corresponds to the sum of the first mismatch result and the second mismatch result.

Accordingly, since only potential target sequences of candidates finally selected by the candidate sequence selector 114 determine whether the potential off-target sites are actual off-target sites (or on-target sites), processing targets to perform mismatch calculation Through the reduction of the performance of the mismatch calculation of the processor 110 (processing speed, processing time, etc.) can be efficient.

4A and 4B are diagrams for explaining the running time of each mismatch number determination when there is no bulge and when there is bulge in the query sequence of sgRNA according to one embodiment.

Referring to FIG. 4A, calculation of mismatch between sgRNA and DNA bound without bulge takes a running time of O (1) according to Big-O notation (uppercase O notation). Here, O (1) is a constant time, and may be the same regardless of the number of mismatched bases. Specifically, the mismatch calculation between two nucleic acid sequences without bulge is based on a bitwise operation and a population count algorithm between binary phenotypes of a nucleic acid sequence, and accordingly the running time is O (1) as described above. ).

However, referring to FIG. 4B, the calculation of a mismatch between the bound sgRNA and DNA due to the occurrence of one bulge (base 'C') is the same as the case without bulge (O (1)). However, since bulges can be present at any position on the nucleic acid sequence of the sgRNA, mismatch calculations must be performed whenever bulges are present at each position on the nucleic acid sequence. Therefore, the running time of mismatch calculation at this time may take as much as O (n) according to Big-O notation. Here, n means the length (ie, nts) of the nucleic acid sequence. Therefore, when bulge occurs, the mismatch calculation time may increase rapidly compared to the case where there is no bulge.

Since the Edit Distance algorithm is used for mismatch calculation between two nucleic acid sequences with bulges, the running time is O (n).

On the DNA, PAM domains can exist in hundreds or tens of millions of places, most of which are candidates for off-target sites. Therefore, assuming that one bulge has occurred in each PAM domain, since each PAM domain takes a running time of O (n), when the number of PAM domains is k, all of k * O (n) Running time may be required.

Therefore, if the number k of target sequences in the off-target site candidate group can be reduced, the total running time can be reduced.

In the field of genetic engineering, a mismatch analysis of a nucleic acid sequence having a bulge in the prediction of an off-target site in the prior art takes a lot of running time (for example, O (n) according to Big-O notation), resulting in a large amount of genome level. Mismatch calculations for sequences were inefficient.

However, according to the present embodiment, a relatively simple and fast dictionary of target sequences having mismatched base number exceeding a user's desired level before full-scale mismatch calculation, except for the binding cases in which two or more bulges, which are very rare cases, exist. By excluding through pre-mismatch processing, it is possible to increase the processing efficiency in a large number of candidate groups. That is, since the value of k described above can be reduced, the processing speed or processing time required for mismatch analysis of the target sequence of the actual off-target site may be reduced.

5 is a diagram for explaining a method of calculating a mismatch between a sgRNA sequence and a DNA sequence in which a bulge is present, according to an embodiment.

Both the sgRNA sequence 510 and the DNA sequence 520 correspond to exemplary arbitrary sequences for convenience of explanation. In addition, in FIG. 5, the bases of the sgRNA sequence 510 and the bases of the DNA sequence 520 are described as performing mismatch analysis using the base symbols A, C, T, G, but the processor 110 is configured to Those skilled in the art can understand that a sequence can be pre-converted to a binary number corresponding to each base symbol, and mismatch analysis between sequences can be performed by comparing binary values.

Referring to FIG. 5, the sgRNA sequence 510 is a nucleic acid sequence in which a bulge of one base A is present, and the total length may be 5nts due to the bulge. DNA sequence 520 is a total of 6nts nucleic acid sequence. That is, assuming that there is no bulge in the sgRNA sequence 510, the number of bases included in each of the sgRNA sequence 510 and the DNA sequence 520 is the same. Meanwhile, the sgRNA sequence 510 of FIG. 5 corresponds to the first query sequence described above, and the DNA sequence 520 corresponds to the target sequence described above.

Mismatch calculation of the sgRNA sequence 510 and the DNA sequence 520 may be performed in a total of 5 digits due to the bulge of the sgRNA sequence 510. Since the mismatch at each digit is output as '11010', the number of mismatch bases corresponding to the mismatch result is 3.

However, as described above, since bulges exist in the sgRNA sequence 510, the number of mismatched bases 3 may be different from the actual mismatch result of the sgRNA sequence 510 and the DNA sequence 520. Accordingly, the sequence analysis apparatus (10 in FIG. 1) according to the present embodiment may perform two steps of pre-processing processes in order to obtain more accurate mismatch results, that is, actual mismatch results.

6 and 7 are diagrams for explaining a first pre-processing process and a second pre-processing process for analyzing actual mismatch results between nucleic acid sequences according to an embodiment.

6 and 7 will be described by taking the nucleic acid sequences of FIG. 5 (sgRNA sequence 510 and DNA sequence 520) as an example for convenience of description, but the pretreatments of FIGS. 6 and 7 have nucleic acids having different bases. The same applies to sequences or nucleic acid sequences of different lengths.

Referring to FIG. 6, in the first pre-processing process, first, the mismatch determination unit 112 of the processor 110 exists in the first query sequence in which the bulge exists (the sgRNA sequence 510 shown in FIG. 5). The first query sequence (the sgRNA sequence 510 shown in FIG. 5) is used as the second query sequence 610 in which the base of the bulge is inserted between the bases G and T at both ends of the bulge in order to assume a non-sequence sequence. To convert.

The mismatch determination unit 112 of the processor 110 obtains a first mismatch result in which the base positions of the converted second query sequence 610 and the target sequence (DNA sequence 520) are matched. The first mismatch result is between the bases located on the first side of the base of the bulge included in the second query sequence 610 and the bases of the target sequence (DNA sequence 520) while the base positions are identical. Corresponds to the number of mismatch bases.

Specifically, the mismatch determination unit 112 has the number of mismatches (m1) of the sequence TG of the first side (right in FIG. 6) of the base A 615 of the bulge in the second query sequence (610, sgRNA (ACGATG sequence)) ) And the number of mismatches (k1) of the sequence ACGA of the second side (left in FIG. 6) including base A 615 of the bulge. At this time, the running time of mismatch calculation in the first preprocessing process will be O (1).

As a result of mismatch calculation in the first preprocessing process, m1 = 1 and k1 = 3 may be obtained. Here, the first mismatch result means the value of m1, and thus the first mismatch result (m1) is 1.

Next, referring to FIG. 7, in the second pre-processing process, the mismatch determination unit 112 of the processor 110 firstly determines the second query sequence 610 for the target sequence (DNA sequence 520). (Right in FIG. 7) is shifted to a state that is misaligned by one base. Then, the processor 110 obtains a second mismatch result in a state where base sites are mismatched. The second mismatch result corresponds to the number of mismatch bases between bases located on the second side of the base of the bulge in the second query sequence 610 and bases of the target sequence (DNA sequence 520).

Specifically, the mismatch determination unit 112 moves the second query sequence 610 (sgRNA (ACGATG sequence)) by one base position in consideration of one bulge. Here, the movement is relative, and the target sequence (DNA sequence 520, CCAGAG sequence) may be moved to the second side (left in FIG. 7).

Referring to FIG. 7, the mismatch determination unit 112 is the second side (left in FIG. 7) sequence ACG of the base A 615 of the bulge in the mismatched second query sequence (610, sgRNA (ACGATG sequence)) The number of mismatches (m2) and the number of mismatches (k2) of the sequence AT of the first side (right in FIG. 7) including base A 615 of the bulge is calculated. At this time, the running time of mismatch calculation in the second pre-processing process will also be O (1).

As a result of mismatch calculation in the second preprocessing process, m2 = 2 and k2 = 1 may be obtained. Here, the second mismatch result means the value of m2, so the second mismatch result (m2) is 2.

That is, the mismatch determination unit 112 of the processor 110 obtains a first mismatch result (m1) and a second mismatch result (m2) as a preprocessing result for determining the actual number of mismatch bases.

The equation that can be derived from the first and second preprocessing processes is equal to 0≤m1 + k1 + m2 + k2-n≤ (n-1). Here, since 0≤k1 + k2≤n, 0≤m1 + m2≤ (n-1).

The candidate sequence selector of the processor (110 of FIG. 1) (114 of FIG. 1) obtains the obtained first mismatch result (m1) and the second mismatch result (m2) and the predefined mismatch threshold number (T). By comparison, it is determined whether to select a target sequence (DNA sequence 520) as a candidate for determining whether it is an on-target site (or an off-target site).

Here, the predefined mismatch threshold number (T) corresponds to a value predefined by the user in order to filter only the number of mismatch bases that is T or less among the off-target site candidate groups in which mismatch has occurred. Accordingly, the candidate sequence selector 114 compares the value of the value (m1 + m2) and the value of T, which is the sum of the first mismatch result (m1) and the second mismatch result (m2), and m1 + m2 ≤ T Only a target sequence that satisfies the condition of is selected as a candidate for determining whether it is an on-target site (or an off-target site).

As a result, the candidate sequence selector 114 selects only a target sequence that satisfies the above condition (m1 + m2≤T) among target sequences of a plurality of potential off-target sites, and thus described in FIGS. 4A and 4B. The value of k can be reduced, and accordingly, the total running time for the processing of the candidate group where the bulge has occurred can be reduced.

Meanwhile, the directions of the first side and the second side described in FIGS. 6 and 7 are opposite directions to each other in an actual nucleic acid sequence. Specifically, when the first side is a 5 'direction of the nucleic acid sequence, the second side is 3 of the nucleic acid sequence. In the 'direction, when the first side is the 3' direction, the second side may be the 5 'direction.

FIG. 8 is a diagram illustrating that a candidate group for determining an on-target site (or for determining an off-target site) may be reduced by preprocessing performed in FIGS. 6 and 7 according to an embodiment. .

Referring to FIG. 8, among the target sequences of all potential off-target sites, a pre-processing process for pruning (removing) target sequences that do not satisfy the conditions of m1 + m2 ≦ T described in FIGS. 6 and 7 above By performing, candidate groups for determining an on-target site (or for determining an off-target site) can be reduced.

When the total number of PAM domains on the DNA (that is, the number of target sequences in the candidate group) is k, the running time of k * O (n) was taken in consideration of bulge, but the number of k is reduced according to the present embodiment. Because it can be, faster on-target site (or off-target site) discrimination is possible.

Table 1 below is a table for explaining the effect of the running time implemented by the present embodiment.

system Processing time per query sequence This Example (CPU only) 0.4 sec Cas-OFFinder (CPU only) 60.0 sec Cas-OFFinder (GPU) 3.01 sec

The present embodiment may be widely used in genetic engineering using a binding principle between short nucleic acid sequences, particularly in the case of mismatch calculation for a large number of comparison candidates at the level of big data. That is, a technique for improving the speed of mismatch calculation between nucleic acid sequences may be essential in big data such as genomes in which a large number of candidates for comparison exist. The sequence analysis device according to the present embodiment (10 in FIG. 1) considers bulge, which is an important factor in mismatch calculation, and particularly, even when there is one bulge, the speed of mismatch calculation is performed through an efficient pruning procedure (pre-processing process). Can be improved.

In Table 1, in addition to the Cas-OFFinder (CPU only) case considering one bulge, in the Cas-OFFinder (GPU) case considering one bulge, an attempt was made to improve speed using a GPU, but the calculation was relatively slow despite the use of a GPU. Speed. When the sequence analysis apparatus 10 according to the present embodiment includes one bulge, it improves the calculation speed through an efficient pruning procedure, so that it is remarkable than the case of using only the CPU and using the GPU of the Cas-OFFinder (GPU) case. It shows a significant improvement in calculation speed.

9 is a flowchart of a method for analyzing a nucleic acid sequence according to an embodiment.

Referring to FIG. 9, the nucleic acid sequence analysis method is composed of steps that are processed in time series in the sequence analysis device 10 of the above-described drawings. Therefore, even if it is omitted below, the contents described with respect to the sequence analysis device 10 of the above-described drawings can also be applied to the method of FIG. 9.

In step 901, the processor 110 obtains information of a first query sequence in which a bulge exists and information of a target sequence of a potential off-target site.

In step 902, the processor 110 converts the first query sequence into a second query sequence in which the base of the bulge is inserted between bases at both ends of the bulge in order to assume that the first query sequence is a sequence in which the bulge does not exist.

In step 903, the processor 110 obtains a first mismatch result in which the base positions of the converted second query sequence and the target sequence are matched and a second mismatch result in which the base positions are mismatched.

In step 904, the processor 110 compares the obtained first and second mismatch results with a predefined number of mismatch thresholds, thereby determining whether the target sequence is an on-target site or an off-target site. It is determined whether to select as a candidate for cognitive discrimination.

Meanwhile, the above-described method may be implemented as a program executable on a computer, and may be implemented on a general-purpose digital computer that operates the program using a computer-readable recording medium. In addition, the structure of data used in the above-described method may be recorded on a computer-readable recording medium through various means. The computer-readable recording medium includes a storage medium such as a magnetic storage medium (eg, ROM, RAM, USB, floppy disk, hard disk, etc.), optical reading media (eg, CD-ROM, DVD, etc.). do.

Those of ordinary skill in the art related to the present embodiment will understand that it may be implemented in a modified form without departing from the essential characteristics of the above-described substrate. Therefore, the disclosed methods should be considered in terms of explanation, not limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the equivalent range should be interpreted as being included in the present invention.

Claims (18)

In the method of analyzing a nucleic acid sequence using a sequence analysis device,
Acquiring information of a first query sequence in which a bulge exists and information of a target sequence of a potential off-target site from a memory by a processor provided in the sequence analysis device;
Converting the first query sequence by the processor into a second query sequence in which the base of the bulge is inserted between the bases at both ends of the bulge in order to assume that the first query sequence is a sequence in which the bulge does not exist. step;
Obtaining, by the processor, a first mismatch result in which the base positions of the converted second query sequence and the target sequence are matched and a second mismatch result in which the base positions are mismatched;
The processor determines whether the target sequence is an on-target site or an off-target site by comparing the obtained first and second mismatch results with a predefined number of mismatch thresholds. And determining whether to select as a candidate for doing so. According to claim 1,
Wherein the bulge is one in the first query sequence. According to claim 1,
The first mismatch result is
With the base positions coinciding, the method corresponds to the number of mismatched bases between bases located on the first side of the base of the bulge included in the second query sequence and bases of the target sequence. The method of claim 3,
The base sites are inconsistent
The second query sequence with respect to the target sequence corresponds to a state that is misaligned by one base to the first side,
The second mismatch result is
A method corresponding to the number of mismatched bases between bases located on the second side of the base of the bulge and bases of the target sequence in the second query sequence. The method of claim 4,
When the first side is the 5 'direction of the target sequence, the second side is the 3' direction of the target sequence,
If the first side is the 3 'direction, the second side is the 5' direction. According to claim 1,
The nucleic acid sequence comprises a plurality of potential off-target sites,
The determining step
By comparing the first and second mismatch results and the number of mismatch thresholds for the target sequence of each of the plurality of potential off-target sites, the on among the target sequences of the plurality of potential off-target sites -Screening the candidate for determining whether it is a target site or the off-target site. The method of claim 6,
The screening step
Selecting the candidate by pruning target sequences having a mismatch base number exceeding the mismatch threshold number among the target sequences of the plurality of potential off-target sites,
The number of mismatch bases
The method corresponds to the sum of the first mismatch result and the second mismatch result. According to claim 1,
The first query sequence
A method corresponding to the nucleic acid sequence of sgRNA for the CRISPR Cas9 system. According to claim 1,
The number of bases included in the target sequence is
Method equal to the number of bases included in the first query sequence. A memory for storing information of a first query sequence in which a bulge exists and information of a target sequence of a potential off-target site; And
The base of the bulge is interposed between bases of both ends of the bulge in order to obtain information of the first query sequence and information of the target sequence from the memory, and to assume the first query sequence as the sequence in which the bulge does not exist Convert the first query sequence to the inserted second query sequence, and the first mismatch result in which the base positions of the converted second query sequence and the target sequence are matched and the base positions are inconsistent Whether the target sequence is an on-target site or off-by obtaining a second mismatch result of and comparing the obtained first and second mismatch results with a predefined number of mismatch thresholds. And a processor that determines whether to select as a candidate for determining whether it is a target site. The method of claim 10,
The bulge is one in the first query sequence. The method of claim 10,
The first mismatch result is
With the base positions coinciding, the device corresponds to the number of mismatched bases between bases located on the first side of the base of the bulge included in the second query sequence and bases of the target sequence. The method of claim 12,
The base sites are inconsistent
The second query sequence with respect to the target sequence corresponds to a state that is misaligned by one base to the first side,
The second mismatch result is
An apparatus corresponding to the number of mismatched bases between bases located on the second side of the base of the bulge and bases of the target sequence in the second query sequence. The method of claim 13,
When the first side is the 5 'direction of the target sequence, the second side is the 3' direction of the target sequence,
When the first side is the 3 'direction, the second side is the 5' direction. The method of claim 10,
The nucleic acid sequence comprises a plurality of potential off-target sites,
The processor
By comparing the first and second mismatch results and the number of mismatch thresholds for the target sequence of each of the plurality of potential off-target sites, the on among the target sequences of the plurality of potential off-target sites A device for selecting the candidate for determining whether it is a target site or the off-target site. The method of claim 15,
The processor
Selecting the candidate by pruning target sequences having a mismatch base number exceeding the mismatch threshold number among the target sequences of the plurality of potential off-target sites,
The number of mismatch bases
The device corresponds to a sum of the first mismatch result and the second mismatch result. The method of claim 10,
The first query sequence
Device corresponding to the nucleic acid sequence of sgRNA for the CRISPR Cas9 system. The method of claim 10,
The number of bases included in the target sequence is
Device equal to the number of bases included in the first query sequence.

Images