[BioPython] Re: BLAST result persistence

ISHIKAWA, Masahiro biopython@biopython.org
03 Jun 2001 09:04:19 +0900


  Hi,

Thanks for your response.

In article <B73C73DA.F85%jchang@smi.stanford.edu>, 
Jeffrey Chang <jchang@SMI.Stanford.EDU> writes:
[snip]
>I'd recommend saving the original blast results, rather than the parsed
>object.  This way, you can always go back to the originals, in case the
>parsed output wasn't what you expected.  When you need it reparse the text
>output.  If you're using a really slow computer, then perhaps you can save a
>pickled cache of the record as well.

I see.
However, in my case, an output file probably contains
results for many number of queries, and it will be more
than several hundreds mega bytes in size.

What I'm worrying about most is how I can handle random
access to each individual result.
I thought plain text doesn't permit me efficient access.

>Plain text file is most convenient for the user.  gzip might be better, if
>you're low on space.

Maybe the whole output plain file cannot be loaded into
memory at a time.
Thus I need some trick to achieve efficient random
access to individual result stored on a disk.

  Thanks again for your advice.