Entering edit mode
10.2 years ago
anlin00007
▴
90
Hello,
I am using tophat to map the ribosome profiling data. First I limited the multiple hits report up to 2, which is:
-g/--max-multihits <int> = 2
Then, I changed this parameter to 5, with no more other changes. However, from the align_summary file, I can see the mapping rates increased, which means more reads mapped to the reference. Based on the tophat manual, this parameter should just change the number of multiple hits in the report which has no effect on the alignment?
I am confusing why the mapping rate changed.
Tophat is one of those programs where a lot is going on under the hood that isn't documented. I wonder if this parameter is getting passed to bowtie2 at some point and that's limiting it to return reads that have at most this number of alignments. If you grep through the BAM file and look at the NH:i:3, NH:i:4 and NH:i:5 alignments and add them up (only counting once per read), is that perhaps the difference in number of aligned reads?
The 'NH:i:3' flags refer to the multiple alignments, not the reads. I am wondering why there are more reads get aligned with no change in the parameters relate to alignment strength.