search

bioperl / bioperl-live-redmine

Legacy tickets migrated from the OBF Redmine issue tracker: http://redmine.open-bio.org
0 stars 0 forks source link

Bio::DB::GenBank/GenPept get_Seq/Stream_by_id no longer retrieves LOCUS IDs #142

Open cjfields opened 8 years ago

cjfields commented 8 years ago

Author Name: Chris Fields (@cjfields) Original Redmine Issue: 3414, https://redmine.open-bio.org/issues/3414 Original Date: 2013-02-19 Original Assignee: Bioperl Guts

Not sure if this is really a bug, as the change is on the NCBI end. My general feeling is we dissuade using this method for retrieving sequences by LOCUS IDs. Note this affects retrieval using SwissProt IDs from NCBI.

Example ones that don’t work anymore: MUSIGHBA1 (GenBank), 2AAA_YEAST (GenPept)

cjfields commented 8 years ago

Original Redmine Comment Author Name: Chris Maloney Original Date: 2013-04-08T01:10:16Z

I’m interested in this issue — I would like to investigate it, but am very new to BioPerl. If I could intercept the HTTP request somehow, I could probably figure out what is wrong (although, fixing it in the BioPerl code would be another matter, of course). Does anybody have any idea about how to do that? If I could change the base URL used by the Bio::DB::GenBank class, and redirect it to my own server, that would do it, but I have no idea how to set that URL.

cjfields commented 8 years ago

Original Redmine Comment Author Name: Chris Maloney Original Date: 2013-04-13T04:55:02Z

I see you fixed this with this commit: https://github.com/bioperl/bioperl-live/commit/eb8d5ef0b7e460f7ef70ec513d8a9c428ffc14e3.

I’d like to take out “not always a locus id”, because I think it’s confusing. Locus id is never allowed there anymore. The value must either be UID or accession number (with or without version).

What’s the best way for me to fix that? It is very minor, but worthwhile, IMO. Should I fork the repo and send a pull request?