[bip] Blog post on bioinformatics and Python

[Andrew Dalke]

On Sep 18, 2008, at 6:27 PM, C. Titus Brown wrote:
> OTOH, frustration with the opaqueness of BioPython BLAST objects (and
> the use of crmpd 'sbjct' vrbl nms) led me to write my own blastparser


The BLAST parser in Biopython was originally written for use
by those who had pretty much memorized how the text output looks.


 >P26583 HIGH MOBILITY GROUP PROTEIN HMG2 (HMG-2).
            Length = 208

  Score =  284 bits (718), Expect = 1e-76
  Identities = 132/168 (78%), Positives = 142/168 (83%)

Query: 1    
GKGDPKKPRGKMSSYAFFVQTCREEHKKKHPDASVNFSEFSKKCSERWKTMSAKEKGKFE 60
            GKGDP KPRGKMSSYAFFVQTCREEHKKKHPD+SVNF+EFSKKCSERWKTMSAKEK KFE
Sbjct: 1    
GKGDPNKPRGKMSSYAFFVQTCREEHKKKHPDSSVNFAEFSKKCSERWKTMSAKEKSKFE 60



BLAST uses "sbjct" and so did the parser.


I'm not saying it was right or wrong, only pointing out the history.


				Andrew
				dalke at dalkescientific.com