[Biopython] SeqIO.index for csfasta files memory issues

Peter biopython at maubp.freeserve.co.uk
Wed Jan 20 11:14:26 UTC 2010


On Tue, Jan 19, 2010 at 11:43 PM, Kevin <aboulia at gmail.com> wrote:
>
> Hi Peter
> It's 64 bit centos shared cluster

OK, good.

> I assumed all the rest of Python and such are the same as well
> but I may be wrong.

It would be worth checking out - if the Python installed is just 32bit,
then hitting a memory limit at 4GB would make sense.

> It's version 1.53 I believe for biopython

OK, good.

> I wanted random access as I need half the reads separated this way
> and I think it is faster. Guess I have to do it the old way.

Could you show us a sample of the data - say just the first 20 reads?
I could then generate a large test file in a similar style to see what
happens if I try and index it on my machine. It would also be nice if
you would allow us to use the sample for a Biopython unit test.

Thanks,

Peter



More information about the Biopython mailing list