kantale
commented
5 years ago Hi, yes I agree. In general, MutationInfo as it is now, is quite outdated. I believe VariantValidator does much of the work set out for MutationTool and is in a great extend based on the hgvs package. I do believe there is still a need for a tool that also supports DbSNP-RSIDs, gene names as reference transcripts in HGVS (although this is discouraged, many do so) and also not-common variation (inversions, conversion, repeats, ...)
I will leave a comment in the first page (README) so that people get warned.
Thanks for the heads up!
As the original author of CoordinateMapper.py (see this post from 2010), I recommend against using it.
That code was written long ago and posted in a biopython mailing list as a sketch. The many discussions about its shortcomings have been lost along the way. For example, the code assumed that alignments were ungapped and that the CDS start was in the first exon, and it would silently give the wrong answer if so. I don't know whether those limitations remain. Although Lena improved it, to my knowledge it has few tests.
Furthermore, it was written at the time when there was literally no publicly available code to perform similar operations. The state of available tools is now much better. I think the hgvs package should be able to do everything you seek from this code.
And, congratulations on MutationInfo... anything that makes variation data easier to use is a good thing (provided it's accurate!).