Decrease in number of multimapping reads after increasing outFilterMultimapNmax and winAnchorMultimapNmax

Skip to first unread message

Kelly H

May 13, 2022, 2:43:55 PMMay 13
to rna-star
Hi Alex,

I am working with Phytophthora organisms and it has been found that they have a lot of repeatability in their genomes. I was interested in changing the the multimapping parameters to increase the number of multimapping reads allowed.  I used:

--outFilterMultimapNmax 200
--winAnchorMultimapNmax 200

I found that in some cases there was a decrease in the number of reads mapped to multiple loci, which is not what I expected and was hoping you might be able to explain why that happened.

For example with a cutoff of 10 for the above values I had the following result:
# reads mapped to multiple loci = 3,204,406
# reads mapped to too many loci = 7,488

With the cutoff of 200
# reads mapped to multiple loci = 3,191,047
# reads mapped to too many loci = 0

So even though 0 reads were considered mapped to too many loci, overall I have less reads because there were less reads mapped to multiple loci.  

3,204,406 + 7,488 - 3,191,047 = 20,847
What happened to the other ~20k reads?

Thank you,

Alexander Dobin

May 13, 2022, 2:52:46 PMMay 13
to rna-star
Hi Kelly,

could you please post the entire Log.out file files, so we can see what happened to other stats.


Kelly Hrywkiw

May 19, 2022, 2:43:59 PMMay 19
to rna-star
Hi Alex, 
Here are the files, Default has the standard cutoff of 10, and in 200Cutoff I set both parameters mentioned earlier to 200.
Thank you!
Reply all
Reply to author
0 new messages