Entering edit mode
7.2 years ago
Chen Sun
★
1.1k
Read quality score is an important information, if the quality score for certain position is very low, then the nucleotide at that position might be a sequencing error. So mapping tools might can not directly use that nucleotide to map, but try to solve the sequencing error issue.
From the paper I read, I did not see any discussion of mapping tools taking advantage of read quality score.
Have anyone seen related information or paper?
Thanks!
Bowtie, bowtie2 and hisat2 all have options to set the input read quality encoding, so they should do something with base quality. Try reading the papers, if the papers do not contain the information, try reading the source from one of them.
thanks, I will try to dig deeper.
Thats why we filter the data before aligning. SNP callers can distinguish sequencing errors from real SNPs
Using quality scores in alignment does not improve alignments, in my testing, and it can incur other problems like different reads calling homopolymer indels at different locations instead of keeping them all left-aligned. It can be useful to trim very low quality bases (such as Q2) prior to alignment, though.