Unexpected Unaligned Single-End Reads

[Dario Strbenac]

I ran into the issue of having reads matching perfectly to hg38 in the unmapped mate1 file when aligning only the individual mate after paired-end alignment. I took one read and put it into its own file and noticed that  % of reads unmapped: other was 100%. I found the reason in an old GitHub issue. By default  --winAnchorMultimapNmax is 50 and BLAT showed that the read had 56 matches to hg38. Perhaps repetitive sequences need their own section in the user manual because the parameter is only shown in Section 15.22 but there's no informative discussion about it.

[Alexander Dobin]

Hi Dario,

thanks for the suggestion, I will add the description of this option into the manual.

Cheers

Alex