formatdb problem in blast 2.2.20 and 2.2.21

Bug #586219 reported by Jose
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
ncbi-tools6 (Ubuntu)
Fix Released
Undecided
Unassigned

Bug Description

Binary package hint: ncbi-tools6

My versions:

Ubuntu 10.04 x86_64

> apt-cache policy blast2
blast2:
  Installed: 1:2.2.21.20090809-1
  Candidate: 1:2.2.21.20090809-1
  Version table:
 *** 1:2.2.21.20090809-1 0
        500 http://ch.archive.ubuntu.com/ubuntu/ lucid/universe Packages
        100 /var/lib/dpkg/status

formatdb runs fine, logs say everything is fine, all index files are output, all fine except for the size of some of the index files differing from older versions of formatdb.

Then when running blastall against the formatted db the output is wrong, I get hits reported but no subject lines in the alignments and 0% identity in all alignments. Also the description of the sequence (though formatdb was run with "-o T") is "unknown".

The same blastall version run against a database formatted with an earlier version of formatdb works ok, so it is not a problem of blastall but of formatdb.

This is an excerpt of the incorrect output:

----
BLASTP 2.2.21 [Jun-14-2009]

Sequences producing significant alignments: (bits) Value

sp|Q51739|AOR_PYRFU Unknown 1234 0.0
sp|Q5JE15|AOR_PYRKO Unknown 1127 0.0
sp|Q9V035|AOR_PYRAB Unknown 1019 0.0
sp|O58778|AOR_PYRHO Unknown 1019 0.0
sp|Q56303|FOR_THELI Unknown 357 2e-97
sp|P76192|YDHV_ECOLI Unknown 191 1e-47
sp|Q58585|Y1185_METJA Unknown 45 0.002

 Score = 1234 bits (3192), Expect = 0.0, Method: Compositional matrix adjust.
 Identities = 0/605 (0%), Positives = 0/605 (0%)

Query: 1 MYGNWGRFIRVNLSTGDIKVEEYDEELAKKWLGSRGLAIYLLLKEMDPTVDPLSPENKLI 60

Query: 61 IAAGPLTGTSAPTGGRYNVVTKSPLTGFITMANSGGYFGAELKFAGYDAIVVEGKAEKPV 120

Query: 121 YIYIKDEHIEIRDASHIWGKKVSETEATIRKEVGSEKVKIASIGPAGENLVKFAAIMNDG 180

Query: 181 HRAAGRGGVGAVMGSKNLKAIAVEGSKTVPIADKQKFMLVVREKVNKLRNDPVAGGGLPK 240

Query: 241 YGTAVLVNIINENGLYPVKNFQTGVYPYAYEQSGEAMAAKYLVRNKPCYACPIGCGRVNR 300

Query: 301 LPTVGETEGPEYESVWALGANLGINDLASIIEANHMCDELGLDTISTGGTLATAMELYEK 360

Query: 361 GHIKDEELGDAPPFRWGNTEVLHYYIEKIAKREGFGDKLAEGSYRLAESYGHPELSMTVK 420

Query: 421 KLELPAYDPRGAEGHGLGYATNNRGGCHIKNYMISPEILGYPYKMDPHDVSDDKIKMLIL 480

Query: 481 FQDLTALIDSAGLCLFTTFGLGADDYRDLLNAALGWDFTTEDYLKIGERIWNAERLFNLK 540

Query: 541 AGLDPARDDTLPKRFLEEPMPEGPNKGHTVRLKEMLPRYYKLRGWTEDGKIPKEKLEELG 600

Query: 601 IAEFY 605

----

This happens for blast 2.2.20 and 2.2.21. I tested 2.2.22 and 2.2.23 and all seems fine. I also tested 32 bit binaries of same blast versions (in a 32bit machine) and they have also the same issue.

Revision history for this message
Launchpad Janitor (janitor) wrote :

This bug was fixed in the package ncbi-tools6 - 6.1.20100808-2

---------------
ncbi-tools6 (6.1.20100808-2) unstable; urgency=low

  * Reupload to unstable following the release of Debian 6.0 (squeeze).

ncbi-tools6 (6.1.20100808-1) experimental; urgency=low

   * New upstream release; the upgrade to BLAST 2.2.24 (from 2.2.21) should
     address a formatdb regression. (LP: #586219.)
   * Uploading to experimental because squeeze is frozen.
   * debian/rules (install-indep-stamp): adapt to new ncbilogo.ico contents.
   * debian/lib{ncbi6,vibrant6a}.symbols: update for latest release.
   * doc/man/*.1: update for latest release.
  [plus the cleanups backported to unstable in 6.1.20090809-2]
 -- Ubuntu Archive Auto-Sync <email address hidden> Sat, 30 Apr 2011 13:21:43 +0000

Changed in ncbi-tools6 (Ubuntu):
status: New → Fix Released
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.