2 Jan 2009 19:38
WWW.qblast
Peter <biopython <at> maubp.freeserve.co.uk>
2009-01-02 18:38:07 GMT
2009-01-02 18:38:07 GMT
On Fri, Jan 2, 2009 at 6:01 PM, Jessica Grant <jgrant <at> email.smith.edu> wrote: > I wrote a script that uses NCBIWWW.qblast and it worked last time I tried it > (a few weeks ago) but this morning I get the following error message: > > File "tblastn.py", line 41, in tblastn > result_handle = NCBIWWW.qblast("tblastn", "nr", fas.seq.data) > File "/root/biopython-1.49/build/lib.linux-i686-2.4/Bio/Blast/NCBIWWW.py", > line 770, in qblast > File "/root/biopython-1.49/build/lib.linux-i686-2.4/Bio/Blast/NCBIWWW.py", > line 837, in _parse_qblast_ref_page > ValueError: A non-integer RTOE found in the 'please wait' page, '' It looks like an empty string was found for RTOE (which obviously cannot be turned into an integer). [As an aside, there was a trivial error in the error processing in Bio/Blast/NCBIWWW.py as it should have said "No RTOE found in the 'please wait' page." instead.] This suggests the NCBI sent back an error page of some kind (or even an empty page if there was some network problem), instead of the normal "please wait" page. Unfortunately, these are not so easy to deal with automatically. Does this happen repeatedly? It would help if you could include the sequence you are trying - then we can attempt to reproduce the error for ourselves. I've tried a tblastn search from my machine and it worked OK. Also, what version of Biopython are you using (I'd guess Biopython 1.49 beta, or 1.49, based on the error message)? > Does this sound like an ncbi error or like something I should be able to > work around? Thanks for the help! Given you say this script used to work, it could be something the NCBI has changed. Have you updated your installation of Biopython since the time the script worked? Thanks, Peter _______________________________________________ BioPython mailing list - BioPython <at> lists.open-bio.org http://lists.open-bio.org/mailman/listinfo/biopython
RSS Feed