[Biopython-dev] New: sequtils.py

Jeffrey Chang jchang at SMI.Stanford.EDU
Mon Jul 30 19:34:28 EDT 2001


Hi Thomas,

>2) we included a quick_fasta_reader hack, the FASTA parser is cool and nice
>    but because of all checkings it takes ages for e.g. a complete genome
>    Should we create a faster alternative ? (compatible with the normal one)

Yes!  It looks like the current Fasta readers are implemented in 
python and might even have some overhead in parsing the description 
lines.  It looks like we should update our FASTA handling to 1) run 
faster and 2) allow optional in-depth parsing of description lines. 
Perhaps it's time to redo it in Martel.



>3) some functions exists in utils.py. Could we move sequence based functions
>    to sequtils.py and use utils.py for other non-seqeunce based functions ?
>    (e.g. I'd like to put my hyper-geometric distribution code there 
>for expression data)

Sure.  However, I'm not sure hyper-geometric distributions should go 
into utils.py.  Perhaps we should start a new package to handle 
statistics and probability-type stuff.


>4) anyone got a hangover from yesterdays banquette ?

Looks like it...  :)

Jeff



More information about the Biopython-dev mailing list