Systems and Methods for Sequence Data Alignment Quality Assessment

Abstract

A computer-implemented method for classifying alignments of paired nucleic acid sequence reads is disclosed. A plurality of paired nucleic acid sequence reads is received, wherein each read is comprised of a first tag and a second tag separated by an insert region. Potential alignments for the first and second tags of each read to a reference sequence is determined, wherein the potential alignments satisfies a minimum threshold mismatch constraint. Potential paired alignments of the first and second tags of each read are identified, wherein a distance between the first and second tags of each potential paired alignment is within an estimated insert size range. An alignment score is calculated for each potential paired alignment based on a distance between the first and second tags and a total number of mismatches for each tag.

What is claimed is: 
     
         1 . A computer-implemented method for classifying alignments of paired nucleic acid sequence reads, comprising:
 receiving a plurality of paired nucleic acid sequence reads, wherein each read is comprised of a first tag and a second tag separated by an insert region;   determining potential alignments for the first and second tags of each read to a reference sequence, wherein each potential alignment satisfies a minimum threshold mismatch constraint;   identifying potential paired alignments of the first and second tags of each read, wherein a distance between the first and second tags of each potential paired alignment is within an estimated insert size range; and   calculating an alignment score for each potential paired alignment based on,
 a distance between the first and second tags, and 
 a total number of mismatches for each tag. 
   
     
     
         2 . The computer-implemented method for classifying alignments of paired nucleic acid sequence reads, as recited in  claim 1 , wherein the paired nucleic acid sequence read is a mate-pair read. 
     
     
         3 . The computer-implemented method for classifying alignments of paired nucleic acid sequence reads, as recited in  claim 1 , wherein the paired nucleic acid sequence read is a paired-end read. 
     
     
         4 . The computer-implemented method for classifying alignments of paired nucleic acid sequence reads, as recited in  claim 1 , wherein the estimated insert size range is a standard deviation of a distribution of estimated insert region sizes for the plurality of paired nucleic acid sequence reads. 
     
     
         5 . The computer-implemented method for classifying alignments of paired nucleic acid sequence reads, as recited in  claim 1 , wherein the calculated alignment score is a function of read alignment length. 
     
     
         6 . The computer-implemented method for classifying alignments of paired nucleic acid sequence reads, as recited in  claim 1 , wherein the calculated alignment score is a function of a total number of possible alignments for each read. 
     
     
         7 . A system for identifying potential alignments for sequencing reads, comprising:
 a nucleic acid sequencer configured to interrogate a sample and produce a plurality of read sequences from the sample; and   a processor in communication with the sequencer, the processor configured to,
 obtain the read sequences from the sequencer, 
 perform alignments of the read sequences from the sequencer to a reference sequence, 
 calculate a quality value for each alignment, and 
 output each alignment with its associated quality value. 
   
     
     
         8 . The system for identifying potential alignments for sequencing reads, as recited in  claim 7 , wherein the quality value is a function of sequencing read type. 
     
     
         9 . The system for identifying potential alignments for sequencing reads, as recited in  claim 8 , wherein the sequencing read is a single fragment read 
     
     
         10 . The system for identifying potential alignments for sequencing reads, as recited in  claim 8 , wherein the sequencing read is a paired read. 
     
     
         11 . The system for identifying potential alignments for sequencing reads, as recited in  claim 10 , wherein aligned paired reads must have insert region sizes that fall within an estimated insert size range for the aligned paired reads. 
     
     
         12 . The system for identifying potential alignments for sequencing reads, as recited in  claim 11 , wherein the estimated insert size range is based on a standard deviation value derived from a distribution of estimated insert region sizes of the aligned paired reads. 
     
     
         13 . A computer-implemented method for determining possible alignments for sequencing reads, comprising:
 interrogating a sample and producing a plurality of read sequences from the sample;   performing alignments of the read sequences from the sequencer;   calculating a quality value for each alignment; and   outputting each alignment with its associated quality value.   
     
     
         14 . The computer-implemented method for determining possible alignments for sequencing reads, as recited in  claim 13 , wherein the quality value is a function of sequencing read type. 
     
     
         15 . The computer-implemented method for determining possible alignments for sequencing reads, as recited in  claim 14 , wherein the sequencing read is a single fragment read. 
     
     
         16 . The computer-implemented method for determining possible alignments for sequencing reads, as recited in  claim 14 , wherein the sequencing read is a paired read. 
     
     
         17 . The computer-implemented method for determining possible alignments for sequencing reads, as recited in  claim 13 , wherein the calculated quality values for each alignment is a function of read alignment length. 
     
     
         18 . The computer-implemented method for determining possible alignments for sequencing reads, as recited in  claim 13 , wherein the calculated quality values for each alignment is a function of number of read mismatches. 
     
     
         19 . The computer-implemented method for determining possible alignments for sequencing reads, as recited in  claim 16 , wherein aligned paired reads have insert region sizes that fall within an estimated insert size range for the aligned paired reads. 
     
     
         20 . The computer-implemented method for determining possible alignments for sequencing reads, as recited in  claim 19 , wherein the estimated insert size range is based on a standard deviation value derived from a distribution of estimated insert region sizes of the aligned paired reads.