Patent Application
20140309945
The present disclosure relates to a sequence alignment apparatus and method, and more particularly, to a sequence alignment apparatus and method capable of forming an alignment permitting all variations and errors that may exist in a read sequence, capable of searching the entire area of a read sequence for variations and errors, and capable of forming an alignment with less computation without permitting backtracking.
Sequence alignment technology is widely used in the entire field of biology. For example, through a process of mapping a read sequence to a known reference sequence, it is possible to complete the genomic sequence of each individual, and moreover, to analyze a variation in sequence between individuals. A large sequencing project, such as the 1000 Genomes Project, is currently under way. When such development continues, it is possible to ultimately provide a personal genome analysis service, a customized medical system according to genetic information, and so on.
The embodiments of the present disclosure are directed to providing a sequence alignment apparatus, method, and program capable of forming an alignment permitting all modifications and errors that may exist in a read sequence and capable of searching the entire area of a read sequence for variations and errors.
The embodiments of the present disclosure are also directed to providing a sequence alignment apparatus, method, and program capable of forming an alignment with less computation without permitting backtracking, unlike existing sequence alignment technology.
According to an aspect of the present disclosure, there is provided a sequence alignment method for aligning a read sequence to a reference sequence, including: searching a reference sequence for a candidate position matched with a fragment, the fragment being a portion of a read sequence; and mapping the read sequence to the reference sequence on the candidate position.
The fragment may be a sequence having a predetermined length from an arbitrary position in the read sequence.
The predetermined length of the fragment may be determined based on a value of an average frequency with which the fragment appears in the reference sequence.
The average frequency may be determined according to a length of the reference sequence and a number of bases.
The searching a reference sequence for a candidate position may include selecting, in the reference sequence, at least one of a position exactly matched with the fragment and a position matched with the fragment within a predetermined error tolerance E.
The searching a reference sequence for a candidate position may include at least one operation of: searching the reference sequence for at least one position exactly matched with the fragment; and performing insertion, deletion, and/or substitution on the fragment within a predetermined error tolerance E, and then searching for at least one position matched with the reference sequence.
The mapping the read sequence to the reference sequence may include mapping a remaining sequence behind the fragment in the read sequence to a sequence behind the candidate position in the reference sequence.
The method may further include determining whether or not the remaining sequence matches with the reference sequence when a portion of the remaining sequence is inserted, deleted and/or substituted with another sequence within the error tolerance E.
The error tolerance E may be an error tolerance set for the reference sequence.
When a portion of the reference sequence behind the candidate position does not match with the remaining sequence behind the fragment in the read sequence, the mapping the read sequence to the reference sequence may include moving a starting position of the reference sequence for matching within the error tolerance E and rematching the remaining sequence to the reference position at the moved starting position.
The method may further include: when the fragment matches with the reference sequence, storing the fragment as a mapping fragment; and when there are portions of the remaining sequence behind the fragment matching with the reference sequence behind the candidate position within the error tolerance E, storing the matched portions as mapping fragments.
The method may further include connecting the mapping fragments to each other when the mapping fragments satisfy the following equation:
|Dr(M1,M2)−DR(M1,M2)|<E−E0
where M1 and M2 are mapping fragments to be connected, Dr(M1, M2) is a distance between the mapping fragments M1 and M2 in a read sequence, DR(M1, M2) is a distance between the mapping fragments M1 and M2 in a reference sequence, E is an error tolerance for the read sequence, E0 is a sum of error values included in the mapping fragments, and |Dr(M1, M2)−DR(M1, M2)| is an absolute value of a difference between Dr(M1, M2) and DR(M1, M2).
According to another aspect of the present disclosure, there is provided a computer-readable medium storing a program for implementing the method described above.
According to another aspect of the present disclosure, there is provided an apparatus for aligning a read sequence to a reference sequence, the apparatus including: a position selector configured to search a reference sequence for a candidate position matched with a fragment, the fragment being a portion of a read sequence; a mapping unit configured to map the read sequence to the reference sequence on the candidate position; and an alignment unit configured to align the read sequence with the candidate position when the reference sequence and the read sequence match with each other on the candidate position.
The fragment may be a sequence having a predetermined length from an arbitrary position in the read sequence.
The predetermined length of the fragment may be determined based on a value of an average frequency with which the fragment appears in the reference sequence, and the average frequency value may be determined according to a length of the reference sequence and a number of bases.
The position selector may be configured to select, in the reference sequence, at least one of a position exactly matching with the fragment and a position matching with the fragment within a predetermined error tolerance E.
The mapping unit may be configured to map a remaining sequence behind the fragment in the read sequence to a sequence behind the candidate position in the reference sequence, or map remaining sequences in front of and behind the fragment in the read sequence to sequences in front of and behind the candidate position in the reference sequence.
The error tolerance E may be an error tolerance set for the reference sequence.
The mapping unit may be configured to determine whether or not the reference sequence behind the candidate position and a remaining sequence behind the fragment in the read sequence matches with each other, and the mapping unit may be configured to move a starting position of the reference sequence for matching within the error tolerance E and rematch the remaining sequence to the reference position at the moved starting position, when a portion of the reference sequence behind the candidate position does not match with the remaining sequence behind the fragment in the read sequence.
The apparatus may further include a storage, wherein the mapping unit may be configured to store, when the fragment matches with the reference sequence, the fragment in the storage as a mapping fragment, and store, when there are portions of the remaining sequence behind the fragment matching with the reference sequence behind the candidate position within the set error tolerance E, the matched portions in the storage as mapping fragments.
The alignment unit may connect the mapping fragments to each other when the mapping fragments satisfy the following equation:
|Dr(M1,M2)−DR(M1,M2)|<E−E0
where M1 and M2 are mapping fragments to be connected, Dr(M1, M2) is a distance between the mapping fragments M1 and M2 in a read sequence, DR(M1, M2) is a distance between the mapping fragments M1 and M2 in a reference sequence, E is an error tolerance permitted for the read sequence, E0 is a sum of error values included in the mapping fragments, and |Dr(M1, M2)−DR(M1, M2)| is an absolute value of a difference between Dr(M1, M2) and DR(M1, M2).
According to one or more exemplary embodiments of the present disclosure, alignment may permit all variations/mutations and errors that may exist in a read sequence, and the entire area of a read sequence may be searched for variations and errors.
In addition, according to one or more exemplary embodiment of the present disclosure, it is possible to form an alignment with less computation without permitting backtracking, unlike existing sequence alignment technology, so that alignment speed may increase.
Exemplary embodiments will now be described more fully with reference to the accompanying drawings to clarify aspects, features, and advantages of the present disclosure. The disclosure may, however, be embodied in many different forms and should not be construed as being limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the present disclosure to those of ordinary skill in the art. It will be understood that when a component is referred to as being “on” another component, the components can be directly on the other component or intervening components.
Also, it will be understood that when an element (or component) is referred to as being operated or executed “on” another element (or component), the element (or component) can be operated or executed in an environment where the other element (or component) is operated or executed or can be operated or executed by interacting with the other element (or component) directly or indirectly.
It will be understood that when an element, component, apparatus, or system is referred to as including a component consisting of a program or software, the element, component, apparatus, or system can include hardware (e.g., a memory or a central processing unit (CPU)) necessary to execute or operate the program or software or another program or software (e.g., an operating system (OS) or a driver necessary for driving hardware), unless the context clearly indicates otherwise.
Also, it will be understood that an element (or component) can be realized by software, hardware, or software and hardware, unless the context clearly indicates otherwise.
The terms used herein are for the purpose of describing particular exemplary embodiments only and are not intended to be limiting. As used herein, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises” and/or “comprising,” when used in this specification, do not preclude the presence or addition of one or more other components.
Hereinafter, the present disclosure will be described in detail with reference to the drawings. In the following description of particular embodiments, many details are provided so as to describe the embodiments in further detail and to aid in understanding the present disclosure. However, those of ordinary skill in the art will appreciate that the embodiments could be used without such details. In some cases, descriptions that are well known but have no direct relationship to the present disclosure will be omitted to prevent the present disclosure from being obscured.
Referring to
The sequencer 10 generates a read sequence from a sample, and the sequence alignment apparatus 100 maps the read sequence generated by the sequencer 10 to a known reference sequence.
The sequence alignment apparatus 100 (referred to as “sequence apparatus 100” below) including the computer-readable recording medium in which the program for performing a sequence alignment method according to an exemplary embodiment of the present disclosure is recorded may perform exact matching based on sequence homology and also inexact matching that permits mismatching within an error tolerance E.
The sequence apparatus 100 according to the present embodiment searches a reference sequence for all mappable positions and determines the mappable positions as candidate positions in consideration of all combinable variations (deletion, substitution, or insertion) for a partial section of the read sequence (referred to as a “fragment” below). Here, the sequence apparatus 100 may search for a position matching with the fragment using a known mapping method (e.g., a method using the Burrows-Wheeler transform (BWT) and a suffix array).
According to an exemplary embodiment of the present disclosure, a start position of the fragment may be determined to be a first base in the read sequence. Alternatively, the start position of the fragment may be determined to be a second base in the read sequence. Alternatively, the start position of the fragment may be determined to be a third base in the read sequence. Alternatively, the start position of the fragment may be determined to be a random position between the first base in the read sequence to a base at half the length of the read sequence. For high accuracy, the position of the fragment is determined to be a section having a predetermined length from the first base of the read sequence, but the present disclosure is not limited to such a position.
Referring to
The sequence apparatus 100 compares a remaining sequence of the read sequence with a reference sequence based on the candidate positions. For example, the sequence apparatus 100 maps a reference sequence R1 right behind the candidate position M1 and the remaining sequence of the read sequence to each other, a reference sequence R2 right behind the candidate position M2 and the remaining sequence of the read sequence to each other, and a reference sequence R3 right behind the candidate position M3 and the remaining sequence of the read sequence to each other.
Meanwhile, when the fragment is not selected from the first position of the read sequence but is selected from any one of subsequent positions, remaining sequences are in front of and behind the fragment. In this case, the sequence apparatus 100 may map a reference sequence right in front of the candidate position as well as a reference sequence right behind the candidate position to the remaining sequences.
When matching is impossible while the sequence apparatus 100 is performing a mapping operation between the remaining sequence of the read sequence and reference sequences of the candidate positions M1, M2, and M3 (e.g., inexact-matching within the error tolerance E is not possible), the sequence apparatus 100 may jump a predetermined distance and then continue to perform the mapping operation. Here, the jump distance may be a value of the maximum error tolerance E according to the sequence length. For example, when the sum of error tolerances of previously selected candidate positions is k, the jump distance may be E−k or less.
Alternatively, when matching is impossible while the sequence apparatus 100 is performing a mapping operation between the remaining sequence of the read sequence and reference sequences, a jump is not performed unconditionally but is performed only if a previous mapping result satisfies a minimum matching distance. Referring to
When a mapping result between the remaining sequence of the read sequence and the candidate position M1 indicates as much matching as the minimum matching length mS or more, the sequence apparatus 100 stores such a matched portion as a mapping fragment (in
When all mapping fragments up to the end of the read sequence are stored, the sequence apparatus 100 attempts to connect the stored mapping fragments. For example, the sequence apparatus 100 determines whether or not mapping fragments are connected based on a read sequence of a mapping fragment, information on a position of the mapping fragment in a reference sequence, and the maximum error tolerance E input as a parameter value.
For example, the sequence apparatus 100 connects mapping fragments when Equation 1 below is satisfied.
|Dr(M1,M2)−DR(M1,M2)|<E−E0 [Equation 1]
Here, M1 and M2 are mapping fragments to be connected,
Dr(M1, M2) is the distance between the mapping fragments M1 and M2 in a read sequence,
DR(M1, M2) is the distance between the mapping fragments M1 and M2 in a reference sequence,
E is an error tolerance for the read sequence,
E0 is the sum of error values included in the mapping fragments, and
|Dr(M1, M2)−DR(M1, M2)| is an absolute value of a difference between Dr(M1, M2) and DR(M1, M2).
The sequence apparatus 100 connects mapping fragments of connectable mapping fragment combinations using a known technique (e.g., the Needleman-Wunsch algorithm) or techniques to be found in the future.
Meanwhile, the length of a fragment may be determined based on the value of an average frequency with which a fragment appears in a reference sequence, and the average frequency value may be determined according to the length of the reference sequence and the number of bases in the reference sequence (i.e., A, G, C, and T). Also, the minimum matching length of mapping fragments may be determined to be the same as the length of a fragment.
Although not shown in the drawings, the sequence apparatus 100 may additionally include hardware and software resources necessary for the program to perform a sequence alignment method according to an exemplary embodiment of the present disclosure. Examples of hardware resources may be a CPU, a memory, a hard disk, and a network card, and examples of software resources may be an OS and a driver for driving hardware. For example, selection of a candidate position or a mapping operation is loaded onto a memory and then performed under the control of a CPU. In this way, to run programs stored in the recording medium 110, hardware resources and/or software resources are necessary. Interaction between these resources and the program stored in the recording medium 110 may be appreciated by those of ordinary skill in the art to which the present disclosure pertains.
Referring to
The position selector 201, the mapping unit 203, the alignment unit 205, and the storage 207 operate in harmony with each other to perform an operation that is the same as or similar to the operation of the sequence apparatus 100 described with reference to
The sequencer 10 generates a read sequence from a sample, and the sequence alignment apparatus 200 maps the read sequence generated by the sequencer 10 to a known reference sequence, thereby aligning the read sequence.
The position selector 201 searches a reference sequence for all mappable positions and determines the mappable positions as candidate positions in consideration of all combinable variations (deletion, substitution, or insertion) for a fragment.
As mentioned above, for high accuracy, the position of the fragment is determined to be a section having a predetermined length from the first base, but the present disclosure is not limited to such a position. In addition, as described in the embodiment of
The mapping unit 203 maps a remaining sequence of the read sequence to the reference sequence based on the candidate positions. Referring to the example of
When matching is impossible while the mapping unit 203 is performing a mapping operation between the remaining sequence of the read sequence and the reference sequences of the candidate positions M1, M2, and M3 (e.g., inexact-matching within the error tolerance E is not possible), the mapping unit 203 may jump a predetermined distance and then continue to perform mapping. Here, the jump distance may be a value of the maximum error tolerance E given to the read sequence or less. For example, when the sum of error tolerances of previously selected candidate positions is k, the jump distance may be E−k or less.
Alternatively, when matching is impossible while the mapping unit 203 is performing a mapping operation between the remaining sequence of the read sequence and reference sequences, a jump is not performed unconditionally but is performed only if a previous mapping result satisfies a minimum matching distance. Referring to
When a mapping result between the remaining sequence of the read sequence and the candidate position M1 indicates as much matchnce as the minimum matching length mS or more, the mapping unit 203 stores such matched portions in the storage 207 as a mapping fragment (in
When all mapping fragments up to the end of the read sequence are stored, the alignment unit 205 connects the stored mapping fragments. For example, the alignment unit 205 determines whether or not mapping fragments are connected based on information on positions of the mapping fragments in the read sequence and the reference sequence, and the maximum error tolerance E input as a parameter value.
For example, when Equation 1 above is satisfied, the alignment unit 205 may connect mapping fragments with respect to connectable mapping fragment combinations using a known technique (e.g., the Needleman-Wunsch algorithm) or techniques to be found in the future.
Referring to
For high accuracy, the position of the fragment may be a first position of the read sequence, but is not limited to the first position. Likewise, the length of the fragment may be determined based on the value of an average frequency with which a fragment appears in a reference sequence so as to increase the speed of sequence alignment, but is not limited to the average frequency value.
The sequence alignment apparatus 100 or 200 maps the fragment selected in step 101 to the reference sequence (S103), and selects candidate positions that exactly match the fragment or match the fragment within an error tolerance (S105).
The sequence alignment apparatus 100 or 200 maps a remaining sequence of the read sequence to the reference sequence based on the candidate positions selected in step 105 (S107).
When mapping is impossible in step 107, the sequence alignment apparatus 100 or 200 may jump a distance within the maximum error tolerance.
The sequence alignment apparatus 100 or 200 connects mapping fragments that satisfy Equation 1 above (S109). In step 109, the sequence alignment apparatus 100 or 200 may fill empty spaces of the mapping fragments using a known technique or a technique to be developed in the future.
A sequence alignment apparatus and method according to the embodiments of the present disclosure described above may be used to search for a single nucleotide polymorphism (SNP), a multiple nucleotide polymorphism (MNP), an indel, an inversion, structural variations, a copy number variation (CNV), etc., and may be used in the entire field of biology, such as in transcriptome analysis and in a determination of a protein binding site for new drug development.
It will be apparent to those skilled in the art that variations can be made to the above-described exemplary embodiments of the present disclosure without departing from the spirit or scope of the present disclosure. Thus, it is intended that the present disclosure covers all such variations provided they come within the scope of the appended claims and their equivalents.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10-2011-0126965 | Nov 2011 | KR | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/KR2012/009981 | 11/23/2012 | WO | 00 | 5/8/2014 |
