MAF to Fasta or XMFA - how does it work?

Question

[Joe]

I'm just about to need to convert some MAF files to fasta alignment files, which from some cursory googling I understand is possible with BioPython/Perl among other tools.

My question is what do the converters actually do to give you the fasta? My understanding at present is that a MAF file contains multiple spans of local sequence alignments, where as an aligned fasta contains sequences that have been gapped until they're the same length.

So when converting, does the resultant fasta simply become all the local alignments concatenated together? Or do you perhaps just get a fasta of every aligned sequence separately?

If it's the latter then I may need a different approach for what I want to do downstream...

Answer 1

[steven.davis]

@jrj.healey This is 15 months after your original post, but may help others. There is a maf2fasta.pl conversion script included with mugsy. How did you solve the conversion problem?

[Joe]

Honestly not sure that I ever did I don’t think. I didn’t trust that I knew what was going on with the MAF format enough to be sure of the downstream results.

I think I switched to using a different program (Kalign or LASTAL if I recall), which would output directly to the right format.