[Bioperl-l] Re: bug #949

[Jason Stajich]

On Thu, 14 Jun 2001, Roger Hall wrote:

> > The current behavior was that the frame sign (+/-) was dependent on the
> > query strand.  I don't think it is.
> 
> The frame sign is directly parsed from the blast report. It is compared to
> the subject strand. The query strand has no direct relationship to either of
> those (as far as BPLite, Sbjct, HSP, and Similarity are concerned - not
> biologically speaking) - and no relationship to the error msg.
> 

So I think the error is that the query strand should be what we are
verifying in blastx reports, but sbjct strand should be verified in
tblastn reports.  

Sbjct is unrelated to frame in blastx reports.  However in tblastn reports
I think the reverse is true...  So the code was written initially based on
the examples we had which were just tblastn reports.  

So we're going to have to detect the blast type in order to actually
verify this correctly or we can just take the easy route which is NOT try
to validate frame with strand...

> 
> 
> And wait just a sec on the second HSP for the first subject:
> 
> > But for the next HSP of the same Sbjct
> >  Score =  156 bits (391), Expect = 3e-37
> >  Identities = 76/97 (78%), Positives = 83/97 (85%)
> >  Frame = +3
> >
> > Query: 2706 GRARWLTPVIPALWDAKVGISSEVRSSRPAWPTWQNSVFMKNTKISQAWWQAPVIPANWE
> 2885
> >             GRARWLTPVIPALW+A+ G S EVRSSRPAWPTW N V  KNTKIS+AWW+APVIPA  E
> > Sbjct: 1    GRARWLTPVIPALWEAEAGGSPEVRSSRPAWPTWXNPVSTKNTKISRAWWRAPVIPATRE
> 60
> >
> > Query: 2886 AEAGESLESRRQRLQ*AEIVPLHSSLGDKSKTLSQKK 2996
> >             AEAGESLE  R+RLQ AEI PLHSSLG+KS+T SQKK
> > Sbjct: 61   AEAGESLEPGRRRLQXAEIAPLHSSLGNKSETPSQKK 97
> 
> Actual Query START= 2706
> Actual Query END = 2996
> 
> when Actual Query START < END  ->  query strand = 1 (! -1): so says
> HSP::new()
> 
> > query start = 2706
> > query end   = 2996
> > query strand= -1
> >
I should have written query strand= 1 here.  

> > sbjct start = 1
> > sbjct end   = 97
> > sbjct strand= 1
> >
> > Frame = +3
> >

> 
> +1 ~ +3 : No warning is generated for this HSP because the subject strand is
> also positive. HSP warnings are given for the HSP's 1 and 3 (plus more).
> 
You're right.  I wasn't being very careful.  oops.

> Please verify!
> 
> Thanks!
> 
> Roger
> 
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l
> 

Jason Stajich
jason@chg.mc.duke.edu
Center for Human Genetics
Duke University Medical Center 
http://www.chg.duke.edu/