Codon change from HGVS.c notation

Question

[asd]

The new version of SnpEff changed the original EFF annotation standard to ANN. The EFF field in the VCF contained the codon change which occured in the cancer sample. The new ANN filed just contains the HGVS.c variant notation.

The new version of SnpEff statistic HTML file has a codon change table, so I guess it's not so complicated to convert the HGVS.c notation to codon change, but I just found the ensembl VEP REST API to do it. How can I convert it in different ways?

For example: ENST00000003084:c.1431_1433delTTC to ccTTCa/cca

[Denise CS]

Not sure what you mean "convert it in different ways". Can you give an example please? Would you like to convert 'NM_015550.3:c.-149-34206G>T' to 'rs123'. This is the documentation for the VEP REST endpoint you may be after.

[asd]

When I wrote "convert it in different ways" I meant "without VEP", because I guess SnpEff doesn't use VEP. I didn't want to use VEP, because I have a lot of data and I don't know the VEP server how much would appreciate me if I send 1000 request per sec.

[dariober]

Loosely related... VEP can be run as a standalone program with a local database so if you have lots of data you don't have do go through their server.

[Denise CS]

I see. Both SnpEff and VEP do the same thing: annotate variants. Whether or not they will appreciate that number of calls, you can find out if you contact them. The latest VEP paper shows a comparison of performance across some variant annotation tools out there.

Answer 1

[Santosh Anand]

SnfEff gives the annotation both at DNA and protein levels.

1. HGVS.c: Variant using HGVS notation (DNA level)
2. HGVS.p: If variant is coding, this field describes the variant using HGVS notation (Protein level). Since transcript ID is already mentioned in ‘feature ID’, it may be omitted here.

The codon information that you are looking for is in the protein level info, which has been now standardized with HGVS notations

[asd]

Thank you for your reply. So using a DNA codon table am I able to convert the HGVS.p to codon change? Are the existing tools for that in R?

[Santosh Anand]

The easiest way out for you is to turn on old behavior using -formatEff flag. See doc ->

Important: This section refers the old annotation format using the 'EFF' tag which can be activated using the -formatEff command line option. As of version 4.1 SnpEff uses the 'ANN' field by default.

Then you can easily merge the two tables in R if you need info from both the old and new format tables

[Denise CS]

The VEP endpoint will give you both codons, the one on the reference genome and the alternative codon.