bowtie 1.2.1.1 --strata reports all valid alignments instead of best stratum #58

Open
ceiabreu opened this Issue Jul 5, 2017 · 0 comments

Comments

Projects
None yet
1 participant

ceiabreu commented Jul 5, 2017

Sorry for the brief example, but the way --strata operates seems to have changed between version 1.2 and version 1.2.1.1:

First, expected behaviour, of version 1.2:

~/src/bowtie-1.2-legacy/bowtie -f -v 1 -a --suppress 2,5,6 --best nHp.2.0 seq1.fa 
seq1	nHp.2.0.scaf00033	485587	0	
seq1	nHp.2.0.scaf04212	22562	0	8:G>T
seq1	nHp.2.0.scaf00720	38269	0	14:T>A
~/src/bowtie-1.2-legacy/bowtie -f -v 1 -a --suppress 2,5,6 --best --strata nHp.2.0 seq1.fa 
seq1	nHp.2.0.scaf00033	485587	0	

(i.e. --strata only reports the first hit, with 0 mismatches)

Now, unexpected behaviour, of version 1.2.1.1:

~/src/bowtie-1.2.1.1/bowtie -f -v 1 -a --suppress 2,5,6 --best nHp.2.0 seq1.fa 
seq1	nHp.2.0.scaf00033	485587	0	
seq1	nHp.2.0.scaf04212	22562	0	8:G>T
seq1	nHp.2.0.scaf00720	38269	0	14:T>A
~/src/bowtie-1.2.1.1/bowtie -f -v 1 -a --suppress 2,5,6 --best --strata nHp.2.0 seq1.fa 
seq1	nHp.2.0.scaf00033	485587	2	
seq1	nHp.2.0.scaf04212	22562	2	8:G>T
seq1	nHp.2.0.scaf00720	38269	2	14:T>A

(i.e. even when --strata is used, all valid alignments are still reported, and they are being treated as the same stratum even though one has a perfect hit and the other two have one mismatch)

Many thanks,

Cei

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment