Entering edit mode
8.6 years ago
ksw
▴
40
Hi,
I am wondering the best way to convert a Multiple Alignment Format file (MAF) generated with Mugsy (http://mugsy.sourceforge.net/) into a VCF format for downstream analysis?
I have looked through past posts and haven't found an answer. It seems that there was an earlier branch of Biopython which could do this but I don't think it is being updated.
If anyone has suggestions, that would be incredible! Thank you!
Did the past posts include this one: converting maf to vcf There is a script linked in that post that is still available to try.
Thanks for your reply. Yes, I looked at the script you posted--it converts files in Mutation Annotation Format (another MAF file, specific to cancer genotyping, I think) to VCF format. The MAF format I am struggling with is Multiple Alignment Format.
A different MAF. What does you file look like (can you show a few lines)? Perhaps this may help: Getting A Vcf File From A Fasta Alignment
Thanks, I'll look into the script you posted. The first lines of the MAF:
##maf version=1 scoring=mugsy a score=640492 label=1 mult=3 s B31.Chromosome 4603 638466 + 910724 AAAATCTCAGGAGAGGGATCAAATGGGGGATGCATAATTCATCCTTCAAGGGTAAGAGACCCAATTACTACTTTACTTAGTATCGTAAAATTATTAAAAATGAAAGAACTTTACCAAATATGGTGTAAATTGTCTAAAAACCACTATAAAGAAAAGTACGACTTAAAAGATATTTTAAATACAACAAATTTTTATAGTAATGTAATAGTATCATCCAAAAAAGCCAATTTAACAAATCTAAAAATAGAAAATCAAGAAATTTTAAAAAGCAATTATGAAAATCTATTGATTAAAGAGATAAAAAGTAATAAGTTATTTCAAGAATTATCAGTAGTTGATTATGAAATCATTAACTATGAAGGCAAAAGACAATCTAAGATTAGAACAGGAGATTCTTCGGGAGGATTAAAGGTATTGT
To update this thread: my issue was that I have called variants using short read data (i.e. generating BAM, then VCF files with respect to a reference genome). I am now interested in including outgroup (FASTA) sequences and need to call variants between the outgroups and the FASTA reference genome. The following tool takes a multiple FASTA alignment (for me, my reference genome and outgroup genomes) and can generate a VCF file:
[https://github.com/sanger-pathogens/snp-sites]
Thanks for your help!