Tweet
Hi everybody,
I'm new to the alignment field so sorry if the question sounds ridiculous.
I want to align Chip-Seq data to the Human genome, where in some loci there is haplotype information. for e.g. in a certain region of a chromosome, I can have several variants of the locus.
In this case the default strategy of the aligners to randomly assing a hit to equally good matches will bias the alignment to the variants, causing a loss of signal. I cannot rely on unique matches, since I would lose all signal that is common to the haplotypes.
How should this be handled?
I was imagining a situation where I would allow the aligner to give me all the equally good hits and resolve the situation manually.
What are the parameters in common aligners like bwa or bowtie to do this?
I cannot see an obvious way of doing this with bwa, for example.
Thanks,
Daniel Sobral
European Bioinformatics Institute
I'm new to the alignment field so sorry if the question sounds ridiculous.
I want to align Chip-Seq data to the Human genome, where in some loci there is haplotype information. for e.g. in a certain region of a chromosome, I can have several variants of the locus.
In this case the default strategy of the aligners to randomly assing a hit to equally good matches will bias the alignment to the variants, causing a loss of signal. I cannot rely on unique matches, since I would lose all signal that is common to the haplotypes.
How should this be handled?
I was imagining a situation where I would allow the aligner to give me all the equally good hits and resolve the situation manually.
What are the parameters in common aligners like bwa or bowtie to do this?
I cannot see an obvious way of doing this with bwa, for example.
Thanks,
Daniel Sobral
European Bioinformatics Institute