[Biopython-dev] ipi parser

Jeffrey Chang jeffrey_chang at stanfordalumni.org
Wed May 19 10:46:03 EDT 2004

On May 19, 2004, at 5:50 AM, Brad Chapman wrote:

> Hi Pierre and Jeff;
> Pierre:
>>> I'm trying to use the Swissprot parser to parse IPI. I read that the
>>> parser should have been fixed for IPI however I get an error on date
>>> when I try to parse ipi.HUMAN I get:
> [...]
>>> ValueError: invalid literal for int(): Human
> Jeff:
>> These errors are nearly always due to changes in the formats of the
>> records that occur from time to time.  Do you have a sample file, or
>> accession number, that I can use to see what's going on?
> I took a look at this using the ipi.HUMAN.dat file from
> ftp://ftp.infobiogen.fr/pub/db/ipi/current/ and was able to
> reproduce the error. It looks like the problem was that the DT lines
> are different then expected:


> I updated the SProt parser to handle this and a patch to
> Bio/SwissProt/SProt.py is attached.

The code looks good!  I've run it on Pierre's file (sent in another 
email) and it seems to run through correctly and generate the right 
results.  Thanks a lot!


More information about the Biopython-dev mailing list