Background: Can skip
First, this is a companion post to this one. In the linked post, I asked a similar question, essentially, "how can I generate a .fasta-like sequence file given (unambiguous) genomic coordinates."
It appears that Gatk's FastaAlternateReferenceMaker is a good option for SNVs and simple indels. However, as it states on the page, this does not work for more complicated variant types:
This tool works only for SNPs and for simple indels (but not for things like complex substitutions.
I decided to repost this as a forum post because I have two lingering questions, and I know that they are not totally "solved" despite a fair amount of interest in them (as below).
Let's begin by visiting the HGVS nomenclature website. The problem of how to correctly write the nucleotide sequences based on a variant descriptor - even very articulated ones such as those found in HGVS/ISCN - is not trivial. As a result, I do not necessarily expect that there is a right answer to this; if there were, undoubtedly it would have been incorporated into other tools (beginning with variant callers themselves, probably!) long ago.
The reverse problem, i.e., how do you name a variant found among (let's face it, NGS reads) in a standard, easily understandable way, is likewise not trivial to solve. There are tools that can check validity of HGVS names - mutalyzer comes to mind, among I'm sure many others, but to my knowledge they do not go directly from sequence to name, nor from HGVS/ISCN name to what the resultant .fasta sequence would be, to the best of my knowledge.
Ok, with this and the prior post as background, my question is, "what are the best options out there for these two goals:
- HGVS/ISCN (or .VCF format) --> .fasta (raw nucleotide sequence)
- .fasta or .bam pile ups --> HGVS/ISCN variant naming?
Thanks very much for your time & help.
