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

SearchIO:blastxml does not return correct query name for blastp (2.2.24) #92

Open cjfields opened 8 years ago

cjfields commented 8 years ago

Author Name: Kris McGary (Kris McGary) Original Redmine Issue: 3154, https://redmine.open-bio.org/issues/3154 Original Date: 2010-11-03 Original Assignee: Bioperl Guts

Bio:SearchIO will return the expected query_name when parsing Blast xml output from Blast version 2.2.22 and earlier. However, it returns ‘Query_1’ for xml output from blastp (2.2.24), instead of the expected (for my search) ‘86959’. Looking at the xml code reveals that the xml field that contains the correct name is unchanged, e.g. “86959”. However, a related xml field has changed and now seems to be providing the data: 2.2.24 says, “Query_1” 2.2.22 says, “lcl|1_0”

cjfields commented 8 years ago

Original Redmine Comment Author Name: Kris McGary Original Date: 2010-11-03T16:29:11Z

Upon further tinkering, I don’t think this is necessarily a bug, just unexpected behavior that varies based on the input. I can still get to the value I’m looking for using query_description.

cjfields commented 8 years ago

Original Redmine Comment Author Name: Jason Stajich Original Date: 2011-03-08T23:51:36Z

We need to see if this is a change in how iterated data is showing up

cjfields commented 8 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2011-04-04T16:29:56Z

Any chance of getting some test data for this?

cjfields commented 8 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2011-08-09T18:54:29Z

This may require a little tinkering with the logic previously used for parsing this data out in the builder classes, as simply changing the tags seems to break parsing of older BLAST/BLAST+ XML data.

cjfields commented 8 years ago

Original Redmine Comment Author Name: Chris Fields Original Date: 2011-08-09T20:20:05Z

I recalled this had been reported elsewhere for biopython; found the original report. As a sidenote, this may very well be a bug on NCBI’s end with the latest BLAST+; still awaiting word back on whether this has been confirmed or not.