hgvs name from variant information

Question

[kesume]

I have a .vcf file with regular variant information as : chr, pos, ref, alt, etc. I wanted to find a hgvs c-syntax and p-syntax for every one of them. Now, I used this tool hgvs to do just that. I found a transcript id from chromosome and pos from refSeq .gtf file [Hg19] and entered the genome, transcript, chrom, pos, ref, alt to get a hgvs_name, but it generates a c-syntax.

How can I similarly generate a p-syntax? I think, the tool will provide me with one if I give a relevant 'NP_...' transcript_id. But where do I find that protein-related id from?

Answer 1

[Denise CS]

Use the Variant Effect Predictor, aka VEP. You can input your VCF (among many other data formats), run the VEP and get both c-syntax and p-syntax (named as HGVSc and HGVSp). You will also get the MAFs from both 1000 Genomes and Exac, clinical significance of the variant (if any) and whether or not it maps to a transcription factor binding profile. Check all possible classes of Output by the VEP. The VEP is available via a web interface, Perl script or through the Ensembl REST API.

Answer 2

[Reece]

The hgvs python package has an experimental script to add a genomic HGVS tag to VCF files. See hgvs/misc/experimental/vcf-add-hgvs from the source package. You may extract that HGVS variant from the resulting VCF file and project it onto transcripts and protein sequences using the hgvs package, or modify the script to do. (The script doesn't do this now because each genomic variant might be associated with a large number of transcripts, and therefore lead to many transcript and protein variants.)

For the moment, it only works with GRCh37 (hg19 refers to the same assembly, but GRCh37 is the preferred name).

Answer 3

[zwdzwd]

You probably want to check out TransVar. It generates c-syntax and p-syntax and can take VCF as your input and much more.

http://transvar.readthedocs.io/en/latest/