Method and system for quantifying sequence alignment
The invention includes methods for aligning reads (e.g., nucleic acid reads, amino acid reads) to a reference sequence construct, methods for building the reference sequence construct, and systems that use the alignment methods and constructs to produce sequences. The invention also includes methods and systems for evaluating the quality of the alignment between the reads and the reference sequence construct. The method is scalable, and can be used to align millions of reads to a construct thousands of bases or amino acids long. The invention additionally includes methods for identifying a disease or a genotype based upon alignment of nucleic acid reads to a location in the construct.
1. A system for quantifying an alignment between a sequence read and a reference sequence construct, the system comprising at least one computer hardware processor and at least one non-transitory computer-readable storage medium, wherein the computer-readable storage medium comprises instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform:
obtaining a sequence read from a genetic sample;
creating, in the storage medium, a reference directed acyclic graph (DAG) data structure representing genetic variability in the sequences of different organisms within a species, wherein the reference DAG data structure comprises a plurality of nodes, wherein each node is stored in the storage medium as a string of one or more symbols and a set of parent nodes, the string representing a nucleotide sequence and the set of parent nodes defining a position with respect to other nodes in the reference DAG data structure, wherein the reference DAG data structure further comprises a first node representing a first conserved region of a reference sequence to which the read aligns, the first node connected by edges to a first alternative node comprising a string including a plurality of symbols representing a structural variation of the reference sequence, and a second alternative node representing a second conserved region of the reference sequence;
aligning the sequence read to the reference DAG data structure by:
creating, in the storage medium, a matrix for each of the first node, first alternate node, and second alternate node, each matrix representing a comparison between the sequence read and the string associated with its respective node;
calculating scores between the sequence read and each of the strings associated with the first node, first alternate node, and second alternate node, wherein each score is based at least in part on a prior score, and wherein, for at least one matrix entry comprising a first symbol of the string associated with its node, the prior score is determined based on scores from matrices of predecessor nodes;
identifying an overall score in the calculated scores, the overall score having a position in one of the created matrices; and
backtracking from the position of the overall score to produce an actual match for the sequence read to the reference DAG data structure, thereby aligning the sequence read to the reference DAG data structure;
determining, based on the aligned position of the sequence read, a number of overlapping bases between the sequence read and the first node, and between the sequence read and the first alternative node or the second alternative node;
determining an overlap value based on the number of overlapping bases or amino acids between the sequence read and either the first alternative node or the second alternative node, thereby quantifying the alignment between the sequence read and the reference sequence construct; and
genotyping the genetic sample, based on the quantified alignment of the sequence read, with respect to the genetic structural variation.
2. The system of claim 1 , wherein aligning the sequence read to the reference DAG data structure comprises simultaneously comparing a string of symbols corresponding to the sequence read to strings of symbols corresponding to the first alternative node and the second alternative node.
3. A method for quantifying an alignment between a sequence read and a reference sequence construct, the method comprising using at least one computer hardware processor to perform:
obtaining a sequence read from a genetic sample;
creating, in a non-transitory computer-readable storage medium, a reference directed acyclic graph (DAG) data structure representing genetic variability in the sequences of different organisms within a species, wherein the reference DAG data structure comprises a plurality of nodes, wherein each node is stored in the storage medium as a string of one or more symbols and a set of parent nodes, the string representing a nucleotide sequence and the set of parent nodes defining a position with respect to other nodes in the reference DAG data structure, wherein the reference DAG data structure further comprises a first node representing a first conserved region of a reference sequence to which the read aligns, the first node connected by edges to a first alternative node comprising a string including a plurality of symbols representing a structural variation of the reference sequence, and a second alternative node representing a second conserved region of the reference sequence;
aligning the sequence read to the reference DAG data structure by:
creating, in the storage medium, a matrix for each of the first node, first alternate node, and second alternate node, each matrix representing a comparison between the sequence read and the string associated with its respective node;
calculating scores between the sequence read and each of the strings associated with the first node, first alternate node, and second alternate node, wherein each score is based at least in part on a prior score, and wherein, for at least one matrix entry comprising a first symbol of the string associated with its node, the prior score is determined based on scores from matrices of predecessor nodes;
identifying an overall score in the calculated scores, the overall score having a position in one of the created matrices; and
backtracking from the position of the overall score to produce an actual match for the sequence read to the reference DAG data structure, thereby aligning the sequence read to the reference DAG data structure;
determining, based on the aligned position of the sequence read, a number of overlapping bases between the sequence read and the first node, and between the sequence read and the first alternative node or the second alternative node;
determining an overlap value based on the number of overlapping bases or amino acids between the sequence read and either the first alternative node or the second alternative node, thereby quantifying the alignment between the sequence read and the reference sequence construct; and
genotyping the genetic sample, based on the quantified alignment of the sequence read, with respect to the genetic structural variation.
4. The method of claim 3 , further comprising accepting an alignment having an overlap value greater than or equal to a threshold value.
5. The method of claim 4 , wherein the threshold is 5 or greater.
6. The method of claim 3 , further comprising providing a confidence level for the genotype based upon the overlap value.
7. The method of claim 3 , further comprising diagnosing a disease based upon the genotype.
8. The method of claim 3 , wherein the first node, the first alternative node, or the second alternative node comprises a string of symbols that is at least 50 bp in length.
9. The method of claim 3 , wherein the reference sequence represents a genome.
10. The method of claim 3 , wherein aligning the sequence read to the reference DAG data structure comprises simultaneously comparing a string of symbols corresponding to the sequence read to strings of symbols corresponding to the first alternative node and the second alternative node.
11. The system of claim 1 , wherein the instructions further comprise accepting an alignment having an overlap value greater than or equal to a threshold value.
12. The system of claim 1 , wherein the instructions further comprise providing a confidence level for the genotype based upon the overlap value.
13. At least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform:
obtaining a sequence read from a genetic sample;
creating, in the storage medium, a reference directed acyclic graph (DAG) data structure representing a reference sequence and genetic variation of the reference sequence, wherein the reference DAG data structure comprises a plurality of nodes, wherein each node is stored in the storage medium as a string of one or more symbols and a set of parent nodes, the string representing a nucleotide sequence and the set of parent nodes defining a position with respect to other nodes in the reference DAG data structure, wherein the reference DAG further comprises a first node representing a first conserved region of the reference sequence to which the read aligns, the first node connected by edges to a first alternative node comprising a string including a plurality of symbols representing a structural variation of the reference sequence, and a second alternative node representing a second conserved region of the reference sequence;
aligning the sequence read to the reference DAG data structure by:
creating, in the storage medium, a matrix for each of the first node, first alternate node, and second alternate node, each matrix representing a comparison between the sequence read and the string associated with its respective node;
calculating scores between the sequence read and each of the strings associated with the first node, first alternate node, and second alternate node, wherein each score is based at least in part on a prior score, and wherein, for at least one matrix entry comprising a first symbol of the string associated with its node, the prior score is determined based on scores from matrices of predecessor nodes;
identifying an overall score in the calculated scores, the overall score having a position in one of the created matrices; and
backtracking from the position of the overall score to produce an actual match for the sequence read to the reference DAG data structure, thereby aligning the sequence read to the reference DAG data structure;
determining, based on the aligned position of the sequence read, a number of overlapping bases between the sequence read and the first node, and between the sequence read and the first alternative node or the second alternative node;
determining an overlap value based on the number of overlapping bases or amino acids between the sequence read and either the first alternative node or the second alternative node, thereby quantifying the alignment between the sequence read and the reference sequence construct; and
genotyping the genetic sample, based on the quantified alignment of the sequence read, with respect to the genetic structural variation.
14. The system of claim 11 , wherein the threshold is 5 or greater.
15. The system of claim 1 , further comprising diagnosing a disease based upon the genotype.
16. The system of claim 1 , wherein the first node, the first alternative node, or the second alternative node comprises a string of symbols that is at least 50 bp in length.
17. The system of claim 1 , wherein the reference sequence represents a genome.
18. The system of claim 1 , wherein the genetic variation of the reference sequence represents genetic variation across a species.
19. The system of claim 1 , wherein the string of symbols representing a genetic structural variation is at least 50 bp in length.
20. The system of claim 1 , wherein the aligned position of the sequence read overlaps the first node and the first alternative node.