Last year I wrote a blog post "BLAST XML output needs more love from NCBI", and in the numerous updates to this, tracked the NCBI outreach and then release of BLAST XML 2.
The new output format was included in BLAST+ 2.2.31 as output format 15, without any kind of beta release for user feedback. Later than planned, I was able to give this a try during the Galaxy Community Conference 2015 Hackathon. Sadly the worries voiced on the OBF Bio* mailing lists were well founded.
In part because XML is so verbose, it is nice to be able to parse it as a stream - meaning capturing the output via stdout and Unix pipes. That appears to be "broken". In fact, producing a bundle of XML files using XInclude seems a recipe for trouble.
Here's a little back-story on my latest preprint (based on my email to samtools-devel), which went live on the biology preprint server bioRvix at the end of last week:
SAM/BAM format v1.5 extensions for de novo assemblies.The current version is a terse three pages (trying to meet an "application note" page limit), but nevertheless should clarify the intended usage of these parts of the SAM/BAM specification.
Peter J. A. Cock, James K. Bonfield, Bastien Chevreux, Heng Li.
bioRxiv DOI: 10.1101/020024
This is another brief NCBI BLAST+ bug report blog post, about a regression in BLAST+ 2.2.29 which will be breaking existing pipelines around the world. The problem is a new "feature" which treats an empty query file as an error.