Suggestion: Make `transcript_match=1` the default for the dbNSFP plugin

[dvg-p4]

Given that VEP is a tool for annotating variants with transcript-level data, I struggle to think of a scenario in which you wouldn't want the dbNSFP annotations to be properly split out and matched by transcript. We've recently been bitten by the undesirable default behavior, when we did some large, expensive VEP runs without realizing we'd have to set an additional option for the dbNSFP plugin to output usable annotations--now we'll have to redo those runs, at least in part. I suspect based on the comments in https://github.com/Ensembl/VEP_plugins/issues/421 and similar tickets that we're not alone in this situation.

[olaaustine]

Hi @dvg-p4, Hope you are well? Thank you for your suggestion. We will have discussions in the team and I will update this ticket with resolution about the default behaviour Thank you Ola

[olaaustine]

Hi @dvg-p4, I hope you're doing well.

Following discussions within the team, we've decided not to alter the default behaviour of the plugin at this time. The key reasons are as follows:

• The plugin may be integrated into other existing pipeline structures, and changing default behaviour could introduce unexpected issues for other users.
• The Ensembl transcript set in dbNSFP may not always be in sync with the current Ensembl transcripts, so we prefer to maintain position-based matching as the default approach.

We aim to keep our documentation fully up to date to provide clear guidance on running the plugin and to help prevent any potential issues.

Thank you. Ola.

[olaaustine]

Hi @dvg-p4, I am going to close this issue now. Please feel free to open another issue or reopen this if you have any other issue. Thank you Ola