3

I recently finished mapping an RNAseq run using STAR2.3.0 and noticed the read1 and read2 counts are different, according to samtools flagstat.

The map% is ~80% but the R1 R2 counts are:

R1=139283692

R2=137472495

This usually does not happen for PE sequencing, which makes me a bit nervous to proceed with analysis. What are some potential reasons something like this could occur? I do not have experience with this happening and can't seem to find reasons by searching the web.

d_kennetz
  • 631
  • 5
  • 17
  • Presumably, the actual read counts in your two fastq source files are the same, right? I have no experience with RNAseq but the first thing that comes to mind is a problem in the sequencing step where too many reads were of low quality and were discarded by the aligner. Does that make sense? – terdon Apr 17 '19 at 18:59
  • The actual read counts of the two fastq source files are the same. It makes sense and was a thought of mine. I was wondering if there is a concrete way to check this out? – d_kennetz Apr 17 '19 at 19:01
  • 1
    Use FastQC to compare the overall quality of both fastq files. – benn Apr 18 '19 at 12:32
  • @d_kennetz Does the actual read counts matches R1? Have you checked which of the reads are missing in the bam file? – Kamil S Jaron Apr 29 '19 at 15:59

0 Answers0