Mutalyzer: Multiple hgvs names

Question

[jan]

Hi,

I'm using Mutalyzer : position converter to convert variants based on chromosomal positions into HGVS c. format. Then, I use Mutalyzer : Name checker to check on variants and get the protein change. The problem comes when mutalyzer output several HGVS transcripts and I need to know which transcript resembles the correct gene and also wildtype/longest transcript. I normally use NCBI to check on the correct transcript. This could be done when I have several genes to analyse.I have variants for 700+ genes so checking the correct transcript in NCBI manually is not practical. Any suggestion on how to handle this issue is highly appreciated

[Ram]

Your title needs a bit of work - it is quite off the topic of your question.

Answer 1

[slw287r]

you may try TransVar that supports batch operations TransVar

Answer 2

[Ram]

As far as my experience on this goes, there is no "right gene". There is usually one NG entry (per gene), multiple NM entries based on the number of transcript variants, and possibly multiple NP entries if multiple proteins are coded for.

You can use eutils to automate queries to NCBI, and frame queries to fetch for you the necessary details. I'd recommend you use either biomaRt or reutils so you can query easily and work with the data you fetch.

You can write an R function that evaluates the longest transcript in a vector of transcript names - it will take a bit of programming, but can be done. Remember that while the longest transcript is usually the canonical transcript, it is not the case all the time. You cannot go by name either - sometimes, the transcript variant named "transcript variant 2" might be the principal transcript.