Closed RolandFaure closed 1 year ago
Hi,
Using gap-affine (1,1,1) you have a gap-opening cost. Therefore, one insertion and one deletion score (1+1)+(1+1) = 4
whereas three mismatches have score 1+1+1=3
. Under edit distance (i.e., (1,0,1)), both alignments score 3
.
Let me know if that makes sense to you.
Yes, thank you very much.
Hi Santiago,
Still working with WFA-lib, I noticed a behaviour that I do not understand on wfa::WFAlignerGapAffine. It seems to me that it did not find the optimal alignment in the example below (so I probably misunderstand the parametrization). I thought that
wfa::WFAlignerGapAffine alignerAffine(1,1,1, wfa::WFAligner::Alignment, wfa::WFAligner::MemoryHigh);
was equivalent towfa::WFAlignerEdit alignerEdit(wfa::WFAligner::Alignment, wfa::WFAligner::MemoryHigh);
, but it does not seem to be the case :gives :
Apparently the gapAffine aligner favored 3 mismatches over one insertion and one deletion, which I do not understand. Do you know why it behaves like this ?
Thank you, Roland