Patent Application
20160171151
1. Field of the Invention
The present invention relates to a method for determining a read error of a base sequence.
2. Description of the Related Art
For identifying mutations such as a single nucleotide polymorphism (SNP), substitution of a plurality of bases, insertion or deletion of a base, there is much demand in a study field, a clinical field, or similar field, for example, diagnosis and treatment, identification of virus and bacteria, and analysis on genetic information of livestock or similar information. For example, WO2006/110855 discloses a technique for efficiently identifying a mutation of an SNP or similar mutation by sequencing using a pyrophosphate sequencing technique.
For example, base sequence data acquired by sequencing using the pyrophosphate sequencing technique or similar technique as disclose in WO2006/110855 might contain a read error in which the number of the consecutive same bases is different from the actual number in a portion having the consecutive same bases. It is difficult to determine whether the difference between a reference sequence and a comparative sequence to be compared with this reference sequence is caused by a mutation or caused by a read error in sequencing.
Therefore, it is an object of the present invention to provide a method for determining whether the difference in base sequence is caused by a mutation or caused by a read error.
To achieve the above-described object, according to one aspect of the present invention, a method for determining a read error of a base sequence is a method for determining whether a difference between a first reference sequence and a plurality of first comparative sequences sharing homology with the first reference sequence is caused by a mutation of the first comparative sequence or caused by a read error in sequencing. The method includes: generating respective second comparative sequences by substituting a sequence having a predetermined base number or more of consecutive same bases for a sequence having the predetermined base number of the consecutive same bases in the respective first comparative sequences; generating a second reference sequence by substituting a sequence having the predetermined base number or more of consecutive same bases for a sequence having the predetermined base number of the consecutive same bases in the first reference sequence; calculating a plurality of edit distances after shortening as edit distances of the respective second comparative sequences with respect to the second reference sequence; calculating an average value after shortening as an average value of the plurality of edit distances after shortening; and determining whether the difference is caused by the mutation or caused by the read error based on the average value after shortening.
To achieve the above-described object, according to one aspect of the present invention, the method for determining the read error of the base sequence is a method for determining whether a difference between a first reference sequence and a first comparative sequence sharing homology with the first reference sequence is caused by a mutation of the first comparative sequence or caused by a read error in sequencing. The method includes: generating a second comparative sequence by substituting a sequence having a predetermined base number or more of consecutive same bases for a sequence having the predetermined base number of the consecutive same bases in the corresponding first comparative sequence; generating a second reference sequence by substituting a sequence having the predetermined base number or more of consecutive same bases for a sequence having the predetermined base number of the consecutive same bases in the first reference sequence; calculating an edit distance of the second comparative sequence with respect to the second reference sequence; and determining whether the difference is caused by the mutation or caused by the read error based on the edit distance.
The present invention allows providing a method for determining whether a difference in base sequence is caused by a mutation or caused by a read error.
A description will be given of a first embodiment of the present invention with reference to the drawings.
Each unit in the determination apparatus 1 communicates with one another through the mother board 11. The CPU 12 executes various operations. The main memory 13 includes, for example, a DRAM and temporarily stores information required for the operation in the CPU 12 or similar information. The HDD 14 stores various information including a program related to the operation executed by the CPU 12 and similar information. The input device 15 includes, for example, user interfaces such as a keyboard and a computer mouse related to input from the user. The output device 16 includes, for example, user interfaces such as a display and a printer related to output to the user. The medium I/F 17 is an interface that communicates with a recording medium 101. The communication I/F 18 is an interface for connecting to a network 102.
The data acquiring unit 31 acquires data related to a base sequence from the recording unit 41 that includes the HDD 14, the recording medium 101, and similar member or from the network 42. This data includes a reference sequence and a comparative sequence. The comparative sequence has the difference from the reference sequence. It is necessary to determine whether the difference is caused by a mutation or a read error. The reference sequence acquired by the data acquiring unit 31 is referred to as a first reference sequence. The comparative sequence is referred to as a first comparative sequence.
Regarding the first reference sequence, the sequence shortening unit 32 substitutes the sequence where three or more of the consecutive same bases are arranged for the sequence where two of those consecutive bases are arranged, so as to generate a shortened base sequence. This shortened base sequence is referred to as a second reference sequence. Regarding the first comparative sequence, the sequence shortening unit 32 similarly substitutes the sequence where three or more of the consecutive same bases are arranged for the sequence where two of those consecutive bases are arranged, so as to generate a shortened sequence. This shortened sequence is referred to as a second comparative sequence.
The edit-distance calculating unit 33 calculates the edit distance of each first comparative sequence with respect to the first reference sequence. Similarly, the edit-distance calculating unit 33 calculates the edit distance of each shortened second comparative sequence with respect to the shortened second reference sequence. The error determining unit 34 determines whether the difference between the first reference sequence and the first comparative sequence is caused by a mutation or a read error based on the edit distance calculated by the edit-distance calculating unit 33.
The determination apparatus 1 according to this embodiment performs the following process. For example, in a pyrophosphate sequencing technique using sequencing such as Pyrosequencing (registered trademark) method, a read error in which an error is made in the number of consecutive bases is likely to be contained in the case where there are the consecutive same bases. The problem containing this error is what is called a homopolymer problem. The determination apparatus 1 determines whether the difference between the first reference sequence and the first comparative sequence is caused by the read error related to the homopolymer problem or the presence of a mutation.
A description will be given of the operation of the determination apparatus 1 according to this embodiment.
In this description, for example, the first reference sequence is defined as “AGCCTTTA” (hereinafter referred to as the sequence 1) and the first comparative sequence is defined as “AGCCTTTTA” (hereinafter referred to as the sequence 2). That is, the sequence 1 and the sequence 2 are different in the number of consecutive “T”s.
In step S102, the edit-distance calculating unit 33 of the calculation unit 30 calculates the edit distance of the first comparative sequence with respect to the first reference sequence. Here, the edit distance is the value obtained by comparing the reference sequence with a target sequence to obtain the presence of an insertion, a deletion, or a substitution of the base and by adding “1” for each presence of the insertion, the deletion or the substitution. For example, with respect to the first reference sequence “AGCCTTTA” (sequence 1), the edit distance of the first comparative sequence “AGCCTTTTA” (sequence 2) is 1 because there is one insertion of T.
In step S103, the sequence shortening unit 32 of the calculation unit 30 generates shortened sequences of the first reference sequence and the first comparative sequence by converting the sequence where three or more bases of the consecutive same bases into the sequence where two bases of those consecutive bases. For example, the sequence 1 is converted into “AGCCTTA” (hereinafter referred to as a sequence 3) and the sequence 2 is converted into “AGCCTTA” (hereinafter referred to as a sequence 4). The shortened sequence of the first reference sequence is referred to as the second reference sequence. The shortened sequence of the first comparative sequence is referred to as the second comparative sequence.
In step S104, the edit-distance calculating unit 33 of the calculation unit 30 calculates the edit distance of the second comparative sequence with respect to the second reference sequence. The edit distance of the sequence 4 with respect to the sequence 3 is 0 because both the sequences coincide with each other.
In step S105, the error determining unit 34 of the calculation unit 30 determines whether the difference between the first reference sequence and the first comparative sequence is caused by a mutation or a read error based on the edit distance before the shortening and the edit distance after the shortening. For example, the error determining unit 34 determines that the difference is caused by a mutation when the edit distance before the shortening and the edit distance after the shortening are the same, and determines the difference is caused by a read error when the edit distance before the shortening and the edit distance after the shortening are different. It may be determined that the difference is caused by a read error when the edit distance after the shortening is 0, and it may be determined that the difference is caused by a mutation when the edit distance after the shortening is other than 0. In the example of the sequence 1 and the sequence 2, the edit distance before the shortening is 1 and the edit distance after the shortening is 0. Therefore, it is determined that the difference between the sequence 1 and the sequence 2 is caused by a read error. The calculation unit 30 outputs the determination result to the output unit 43 including the HDD 14, the output device 16, the recording medium 101, the network 102, and similar member so as to, for example, record the determination result. Then, the process terminates.
The read error related to the homopolymer problem hardly occurs even when two of the consecutive same bases appear. When three or more of the consecutive same bases appear, the occurrence rate of the error increases as the number of consecutive bases is larger. Therefore, in this embodiment, when the consecutive same bases appear, all the numbers of consecutive bases are shortened to two such that the analysis is not affected by a read error in sequencing. That is, in the case where there is a read error in sequencing caused by the homopolymer problem, the difference between the sequences is eliminated by shortening. In the case where there is a mutation without a read error, the difference is maintained by shortening. These conditions are used.
Another example of the first comparative sequence will be described. The first reference sequence is defined as the sequence 1, and the first comparative sequence is defined as “AGCCGTTA” (hereinafter referred to as a sequence 5). At this time, the edit distance of the sequence 5 with respect to the sequence 1 calculated in step S102 is 1 because one of “T” is substituted for “G.” The second comparative sequence obtained by shortening the sequence 5 in step S103 is “AGCCGTTA” (hereinafter referred to as a sequence 6). The edit distance of the sequence 6 with respect to the sequence 1 calculated in step S104 is 1. Accordingly, in step S105, it is determined that there is a mutation.
As described above, according to this embodiment, a homopolymer is shortened in the analysis of the sequence data containing the homopolymer problem. Based on the edit distance of the shortened sequence, it is determined that the difference in the sequence is caused by a mutation or a read error. That is, this embodiment allows accurately identifying an SNP or similar mutation.
The processes described with reference to
A second embodiment of the present invention will be described. Here, the difference from the first embodiment will be described. Like reference numerals designate corresponding or identical elements, and therefore such elements will not be further elaborated here. In the sequencing using the pyrophosphate sequencing technique, a large amount of sequence data is obtained at once. In this embodiment, a group of sequences sharing homology with a reference sequence is treated as sequence data. The determination apparatus 1 according to this embodiment determines whether the difference is caused by a read error or by a heterozygous SNP or similar mutation in a portion having the difference from the reference sequence in this group of sequences. Here, the SNP or similar mutation denotes a substitution, an insertion or a deletion of one base. Like the homopolymer problem, in the case where the read error occurs not randomly but under a certain condition, the read error might be determined as the SNP or similar mutation by a conventional determination method. In this embodiment, this read error containing the homopolymer problem is discriminated. Here, in the case where there is a heterozygous SNP or similar mutation, the average value of the edit distances of the respective sequences with respect to the reference sequence ideally becomes 0.5. Accordingly, there are two types of sequences at the respective rates of 50%.
A description will be given of the operation of the determination apparatus 1 according to this embodiment.
In step S202, the edit-distance calculating unit 33 of the calculation unit 30 calculates the edit distances of the respective first comparative sequences with respect to the first reference sequence. For example, the edit distance of the respective sequences shown in Table 1 is calculated as follows. Since the respective sequences 7 to 9 are the same as a consensus sequence, the respective edit distances are 0. In the sequences 10 and 11, since one of T is inserted into the consensus sequence, the edit distance is 1. In the sequence 12 and 13, since one of T is deleted from the consensus sequence, the edit distance is 1. In the sequence 14, since T is substituted for G in the consensus sequence, the edit distance is 1. In step S203, the edit-distance calculating unit 33 of the calculation unit 30 calculates the average value of the edit distances of all the first comparative sequences.
In step S204, the sequence shortening unit 32 of the calculation unit 30 generates shortened sequences of the first reference sequence and the respective first comparative sequences by converting the sequence where three or more bases of the consecutive same bases into the sequence where two of those consecutive bases. For example, the sequences 7 to 14 as the first comparative sequences shown in Table 1 are converted into respective sequences 15 to 22 as the shortened second comparative sequences shown in Table 2.
In step S205, the edit-distance calculating unit 33 of the calculation unit 30 calculates the edit distances of the respective second comparative sequences with respect to the second reference sequence. This edit distances are, for example, as shows in Table 2. In step S206, the edit-distance calculating unit 33 of the calculation unit 30 calculates the average value of the edit distances of the second comparative sequences.
In step S207, the error determining unit 34 of the calculation unit 30 determines whether the difference between the first reference sequence and the first comparative sequence might be caused by an SNP or similar mutation or by the read error based on whether or not the average value of the edit distances of the second comparative sequences is within a predetermined range. For example, the determination is made based on determination criteria shown in Table 3. That is, in the case where the average value of the edit distances of the second comparative sequences (after the shortening) is equal to or more than 0.25 and equal to or less than 0.75, it is determined that an SNP or similar mutation might be the cause. This is based on the fact that the edit distance is ideally 0.5 in the case of a heterozygous diploid in which there is an SNP or similar mutation in one allele. Here, while in this embodiment the range of the average value of the edit distances is set to be equal to or more than 0.25 and equal to or less than 0.75 as the value having 0.5 in the center, another value including 0.5 may be set.
For the reason described later in detail, in the case where the average value of the edit distances of the first comparative sequences (before the shortening) is equal to or more than 0.25 and equal to or less than 0.75 even when the average value of the edit distances of the second comparative sequences (after the shortening) is larger than 0.75, it is determined that an SNP or similar mutation might be the cause. Otherwise, it is determined that a read error is the cause. When it is determined that an SNP or similar mutation might be the cause in step S207, the process proceeds to step S208. When it is determined that a read error is the cause, the process proceeds to step S210.
In step S208, the error determining unit 34 of the calculation unit 30 determines whether or not the proportion of the number of sequences that might have an SNP or similar mutation is within a predetermined range so as to determine whether the difference between the first reference sequence and the first comparative sequence might be caused by an SNP or similar mutation or is caused by a read error. For example, when the number of the sequences that might have an SNP or similar mutation is 33% or more and 67% or less of the number of all sequences, it is determined that the difference is caused by an SNP or similar mutation. Otherwise, it is determined that the difference is caused by a read error. This determination may be changed as follows. When the second comparative sequences are categorized into respective groups of the same sequences, in the case where the number of sequences in the group including the largest number of sequences is 66% or less of the number of all the sequences and the number of sequences in the group including the second largest number of sequences is 33% or more of the number of all the sequences, it is determined that the difference is caused by an SNP or similar mutation.
When it is determined that the difference is caused by an SNP or similar mutation, the process proceeds to step S209. When it is determined that the difference is caused by a read error, the process proceeds to step S210. That is, when the condition of the determination in step S207 and the condition of the determination in step S208 are both satisfied, the process proceeds to step S209. Otherwise, the process proceeds to step S210.
In step S209, the error determining unit 34 of the calculation unit 30 identifies the sequence that having an SNP or similar mutation. The calculation unit 30 outputs information related to the sequence having the SNP or similar mutation to the output unit 43 so as to, for example, record the information. Then, the process terminates.
On the other hand, in step S210, the error determining unit 34 of the calculation unit 30 concludes that the difference between the first reference sequence and the first comparative sequence is caused by a read error and outputs the result to the output unit 43 so as to, for example, record the result. Then, the process terminates.
In the first example of the clusters shown in Table 1, in step S205, the edit distances of the respective second comparative sequences after the shortening are as shown in Table 2. In this case, the average value of the edit distances of the second comparative sequences after the shortening to be calculated in step S206 is 0.125. Accordingly, in the determination in step S207, it is determined that there is a read error. In step S210, a read error is concluded to be present. Then, the process terminates.
In the case of the example shown in Table 1, with this embodiment, the sequence shortening unit 32 shortens the sequence so as to exclude the error caused by the homopolymer problem, and thus concludes that the difference between the first reference sequence and the first comparative sequence is caused by a read error.
Table 4 shows another example of a group of the first comparative sequences. In this case, the edit distances calculated in step S202 are as shown in Table 4. The average value of the edit distances calculated in step S203 is 0.625.
As a result of shortening in step S204, sequences 23 to 30 are changed into respective sequences 31 to 38 as the second comparative sequences shown in Table 5. In this case, the edit distances after the shortening to be calculated in step S205 are as shown in Table 5. The average value of the edit distances after the shortening to be calculated in step S206 is 0.375.
In this case, in the determination in step S207, since the average value of the edit distances after the shortening is equal to or more than 0.25 and equal to or less than 0.75, it is determined that an SNP or similar mutation might be present. Then, the process proceeds to step S208. According to the sequences after the shortening shown in Table 5, sequences 36 to 38 are sequences that might have an SNP or similar mutation. Here, 3 as the number of the sequences is 37.5% of 8 as the number of all the sequences. Accordingly, it is also determined that an SNP or similar mutation is present in step S208. The process proceeds to step S209.
Step S209 identifies the fact that the sequences 28 to 30 corresponding to the sequences 36 to 38 have an SNP or similar mutation in the first comparative sequences shown in Table 4. Like this example, this embodiment allows accurately identifying the sequence having an SNP or similar mutation also in the first comparative sequences having the homopolymer problem.
Table 6 shows another example of the group of the first comparative sequences. In this case, the edit distances of sequences 39 to 46 to be calculated in step S202 are as shown in Table 6. The average value of the edit distances to be calculated in step S203 is 0.625.
As a result of shortening in step S204, the sequences 39 to 46 are changed into respective sequences 47 to 54 as the second comparative sequences shown in Table 7. In this case, the edit distances after the shortening to be calculated in step S205 are as shown in Table 7. The average value of the edit distances after the shortening to be calculated in step S206 is 1.
In this example, in the determination in step S207, the average value of the edit distances after the shortening is equal to or more than 0.75 and the average value of the edit distances before the shortening is equal to or more than 0.25 and equal to or less than 0.75. Accordingly, it is determined that the SNP or similar mutation might be contained. The process proceeds to step S208. According to the second comparative sequences after the shortening shown in Table 7, the sequences 51 to 54 are sequences that might have an SNP or similar mutation. Here, 4 as the number of the sequences is 50% of 8 as the number of all the sequences. Accordingly, it is determined that an SNP or similar mutation is present. The process proceeds to step S209. Step S209 identifies the fact that the sequences 43 to 46 corresponding to the sequences 51 to 54 have an SNP or similar mutation in the first comparative sequences shown in Table 6.
Like this example, in the case where there is a substitution of one base in between the sequence having the consecutive same bases, the edit distance after the shortening becomes considerably large. Accordingly, making the determination whether or not separation is necessary based only on the determination whether or not the edit distance after the shortening is equal to or more than 0.25 and equal to or less than 0.75 causes an incorrect result. Thus, in this embodiment, also in the case where the edit distance after the shortening is equal to or more than 0.75 and the edit distance before the shortening is equal to or more than 0.25 and equal to or less than 0.75, it is determined that separation is necessary. This accurately identifies an SNP or similar mutation.
Table 8 shows another example of the group of the first comparative sequences. In this case, the edit distances of the sequences 55 to 62 to be calculated in step S202 is as shown in Table 8. The average value of the edit distances to be calculated in step S203 is 0.125.
The second comparative sequences as the result by shortening the sequences 55 to 62 in step S204 are changed into respective sequences 63 to 70 shown in Table 9. In this case, the edit distances after the shortening to be calculated in step S205 are as shown in Table 9. The average value of the edit distances after the shortening to be calculated in step S206 is 0.375.
In this case, in the determination in step S207, the average value of the edit distances after the shortening is equal to or more than 0.25 and equal to or less than 0.75. Accordingly, it is determined that an SNP or similar mutation might be present. Then, the process proceeds to step S208. According to the sequences after the shortening shown in Table 9, for the first comparative sequence, the sequence 70 is the sequence that might have an SNP or similar mutation. Here, 4 as the number of the sequences is 12.5% of 8 as the number of all the sequences. Accordingly, it is determined that a read error is present. Then, the process proceeds to step S210. Step S210 concludes that a read error is present. Then, the process terminates.
Like this example, even when the average value of the edit distances after the shortening is equal to or more than 0.25 and equal to or less than 0.75 and it is determined that an SNP might be present in step S207, in the case where the number of sequences is considerably biased, it is determined that a read error is present in the determination in step S208. Accordingly, it can be concluded that the difference is caused by a read error.
Thus, according to this embodiment, in the analysis of the sequence data containing the homopolymer problem, the homopolymer is shortened. Based on the edit distance of the shortened sequence, it is accurately determined whether the difference between the first reference sequence and the first comparative sequence is caused by an SNP or similar mutation or is caused by a read error. That is, this embodiment allows accurately identifying an SNP or similar mutation.
Here, the processes described with reference to
Here, also in case of a heterozygous diploid and in the case where approximately half of the first comparative sequences is different from the reference sequence and contains a first SNP while the remaining sequences contain another second SNP, the determination criteria in step S207 corresponding to Table 3 is, for example, as shown in Table 10. In the determination in step S208, setting is done so as to determine that an SNP or similar mutation is contained in the case where the number of the first comparative sequences containing the first SNP and the number of the second comparative sequences containing the second SNP are each 33% or more and 67% or less of the number of all the sequences and otherwise determine that the difference of the sequence is caused by a read error.
For use other than the identification of an SNP or similar mutation, the configuration with three or more categories of the group by the mutation is possible. In this case, the determination criteria in step S207 and the determination criteria in step S208 become different. For example, in the case of the categorization into four groups, the range of the determination criteria in step S207 is set to the range having 0.25 in the center.
While in this embodiment the edit distance with respect to the reference sequence is obtained, the edit distance with respect to the consensus sequence of a group of sequences may be obtained. Here, the consensus sequence means the sequence determined so as to have the highest identity with respect to all of the group of the first comparative sequences, that is, the sequence in common between a large number of the first comparative sequences.
The calculation unit 30 of the determination apparatus 1 may further include a cluster determining unit 35 as illustrated in
Thus, according to this embodiment, it is determined whether the difference in sequence is caused by a mutation or a read error in sequencing. Using this determination allows determining if variation of genomes or cDNA sequences is true. Additionally, this allows determining if the mutated portion of the cluster generated by assembly of the sequence is true. Additionally, this allows dividing the cluster depending on the determination result, and this can be a pretreatment method for generating clusters.
A third embodiment of the present invention will be described. Here, the difference from the second embodiment will be described. Like reference numerals designate corresponding or identical elements, and therefore such elements will not be further elaborated here. In this embodiment, the method for identifying the SNP or similar mutation according to the second embodiment and a method for generating clusters using this method are applied to a high coverage expression profiling (HiCEP) method. Accordingly, the determination apparatus 1 according to this embodiment has the configuration shown in
The outline of the HiCEP will be described with reference to
With the HiCEP as described above, regarding 30 to 60 thousand types of fragments of the mRNA in the sample, each of information related to the number of molecules can be acquired. The HiCEP allows exhaustively acquiring a highly-reproducible expression profile with high sensitivity without requiring a genetic resource.
For example, when the fragment whose expression varies is detected by the HiCEP, it is required to identify the gene of this fragment. For exhaustive DNA sequencing, a next-generation sequencer at a high throughput is used. When the homopolymer problem that occurs in sequencing particularly using the pyrophosphate sequencing technique in the next-generation sequencer occurs in the selection base portion adjacent to the adaptor sequence, this problem might negatively affect the analysis using the HiCEP. Thus, in this embodiment, the method for identifying the SNP or similar mutation according to the second embodiment is used for the selection base portion.
In step S302, regarding the first consensus sequence and the respective first comparative sequences, the sequence shortening unit 32 of the calculation unit 30 shortens the sequence having three or more bases of the consecutive same bases to a sequence having two bases of the consecutive same bases. The sequences after shortening the sequences shown in Table 11 are shown in Table 12. The sequences 71 to 78 are shortened to respective sequences 79 to 86.
In step S303, the edit-distance calculating unit 33 of the calculation unit 30 calculates the edit distance of the first comparative sequence before the shortening with respect to the first consensus sequence. In step S304, the edit-distance calculating unit 33 of the calculation unit 30 calculates the average value of the edit distances of the first comparative sequences before shortening. In step S305, the edit-distance calculating unit 33 of the calculation unit 30 calculates the edit distance of the second comparative sequence after the shortening with respect to the second consensus sequence generated by shortening the first consensus sequence. In step S306, the edit-distance calculating unit 33 of the calculation unit 30 calculates the average value of the edit distances of the second comparative sequences after the shortening.
Here, for the edit distance, five bases closest to the adaptor sequence in the read sequence and the selection base in the second consensus sequence are set as targets. The edit distances of the second comparative sequences corresponding to these five bases are calculated. That is, the edit distance is calculated regarding the base sequences shown in “Read Sequence and Selection Base” in Table 11 and Table 12. Since the attention is focused on the sequence corresponding to the consensus sequence, the target sequences are not limited to five bases, for example, like the sequences 84 to 86. The edit distances are as shown in Table 11 and Table 12.
In step S307, the error determining unit 34 of the calculation unit 30 determines whether or not the first comparative sequence might contain an SNP or similar mutation corresponding to whether or not the average value of the edit distances is within a predetermined range. For this determination, for example, the determination criteria shown in above-described Table 3 is used. That is, in the case where the edit distance after the shortening is equal to or more than 0.25 and equal to or less than 0.75, it is determined that the first comparative sequence might contain an SNP or similar mutation. Even when the edit distance after the shortening is larger than 0.75, in the case where the edit distance before the shortening is equal to or more than 0.25 and equal to or less than 0.75, it is determined that the first comparative sequence might contain an SNP or similar mutation. Otherwise, it is determined that it is determined that the difference is caused by a read error. When it is determined that the first comparative sequence might contain an SNP or similar mutation, the process proceeds to step S308. When it is determined that the difference is caused by a read error, the process proceed to step S311.
In step S308, when the cluster determining unit 35 of calculation unit 30 assumes the manner of separation when the first base sequence is separated based on two bases on the most adaptor sequence side of the read sequence. This embodiment is focused only on the two bases on the most adaptor sequence side of the read sequence. This is because there is a selection base in this portion and the emphasis is on this selection base. For example, in the example in Table 12, there are two types of “CC” in the sequence 79 to 83 and “CG” in the sequences 84 to 86. Accordingly, the cluster determining unit 35 assumes that the sequences 79 to 86 are separated into two of the cluster including the sequences 79 to 83 and the cluster including the sequences 84 to 86.
In step S309, the error determining unit 34 of the calculation unit 30 determines whether or not the first comparative sequence might contain an SNP or similar mutation corresponding to whether or not the number of sequences included in each cluster is within a predetermined range in the case where the first comparative sequence is separated into two portions. For example, when two clusters are obtained by separation, in the case where the respective numbers of sequences in both the clusters are equal to or more than 33% and equal to or less than 67% of the number of all the sequences, it is determined that the first comparative sequence contains an SNP or similar mutation. This determination may be changed as follows. When the first comparative sequences are categorized into respective groups of the same sequences, in the case where the number of sequences in the group including the largest number of sequences is 66% or less of the number of all the sequences and the number of sequences in the group including the second largest number of sequences is 33% or more of the number of all the sequences, it is determined that an SNP or similar mutation is contained. Otherwise, it is determined that the difference in sequence is caused by a read error. When it is determined that the first comparative sequence contains an SNP or similar mutation, the process proceeds the w step S310. When it is determined that the difference in sequence is caused by a read error, the process proceeds to step S311.
In the example shown in Table 12, the number of sequences in the cluster including the sequences 79 to 83 is 5 that is 62.5% of 8 as the number of all the sequences. On the other hand, the number of sequences in the cluster including the sequences 84 to 86 is 3 that is 37.5% of 8 as the number of all the sequences. Accordingly, it is determined that the first comparative sequence contains an SNP or similar mutation. Then, the process proceeds to step S310.
In step S311, the cluster determining unit 35 of the calculation unit 30 separates the first comparative sequences into the clusters assumed in step S308 so as to generate two clusters. The calculation unit 30 outputs information related to the generated clusters to the output unit 43 so as to, for example, record the information. Then, the process terminates.
In step S311, the cluster determining unit 35 of the calculation unit 30 concludes that difference between the first consensus sequence and the first comparative sequence is caused by a read error. The calculation unit 30 outputs a notification indicating that the difference between the first consensus sequence and the first comparative sequence is caused by a read error to the output unit 43 so as to, for example, record the notification. Then, the process terminates.
This embodiment allows solving the homopolymer problem that become a problem also in analysis using the HiCEP, thus accurately performing clustering in the selection base portion that is important in the HiCEP. Here, this embodiment is not limited to the HiCEP and can be similarly used in another analysis using the adaptor sequence.
A fourth embodiment of the present invention will be described. Here, the difference from the first embodiment will be described. Like reference numerals designate corresponding or identical elements, and therefore such elements will not be further elaborated here. In this embodiment, similarly to the process in step S103 in the first embodiment, a second base sequence is generated. In the second base sequence, the sequence having three or more of the consecutive same bases is shortened to be converted into the sequence having two of those bases. Based on this second base sequence, an already-known clustering process is performed.
This embodiment shortens the sequence having the consecutive same bases to have a short base length. This improves the efficiency of clustering. Additionally, the sequence having the consecutive same bases is shortened so as to solve the homopolymer problem. This improves the accuracy of clustering.
Here, this embodiment is applicable to the sequencing result related to the sample obtained by the HiCEP. In the case of the data related to the HiCEP, the fragment lengths are set to the same length by the adaptor sequences. This allows obtaining a good clustering result, and this embodiment especially provides the effects.
It should be noted that the present invention is not completely limited to the embodiments described above, and modifications of components can be made at the stage of carrying out the invention without departing from the spirit thereof. Further, various inventions can be made by properly combining the components disclosed in the embodiments described above. For example, when the problems described in the section of PROBLEMS TO BE SOLVED BY THE INVENTION can be solved and the advantages of the invention can be obtained even if some of all the components shown in the embodiments are eliminated, a configuration in which those components are eliminated can also be extracted as an invention. Moreover, components in different embodiments may be suitably combined together.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2012-101755 | Apr 2012 | JP | national |
This application is a Continuation Application of PCT Application No. PCT/JP2013/062426, filed Apr. 26, 2013 and based upon and claiming the benefit of priority from prior Japanese Patent Application No. 2012-101755, filed Apr. 26, 2012, the entire contents of all of which are incorporated herein by reference.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/JP2013/062426 | Apr 2013 | US |
| Child | 14523532 | US |
