[Bioperl-l] Bio::SearchDist problem

Jason Stajich jason.stajich at duke.edu
Sat Sep 10 17:11:33 EDT 2005


I see this too.  Ewan will have to report in on what is wrong.  I  
know this code is ancient but it should at least do what it says it  
can do.

Can you go ahead and submit it as a bug to bugzilla.
http://bugzilla.open-bio.org/

Thanks,
-jason
On Sep 10, 2005, at 9:22 AM, Gabor Toth wrote:

> Hi!
>
> I would like to use Bio::SearchDist to fit an extreme value  
> distribution to a
> sorted set of 20724 integer scores and calculate E-value for each  
> score. The
> scores are distributed around a median of zero with a range between  
> -204 and
> +1073.
>
> However, using the script below, I cannot get meaningful E-values.  
> The first
> 29 "E-values" are 1075530112, the 30th is 1, and all other are zero.
> (The input and output files can be found here:
> http://bioweb.abc.hu/tmp/evd/index.html).
> My BioPerl version is 1.4.
>
> What am I doing wrong?
> I will appreciate any help,
>
> Gabor Toth
>
> ------------------------------------------------------
> #!/usr/bin/perl -w
>
> use Bio::SearchDist;
>
> $dis = Bio::SearchDist->new();
>
> while (<>) {
>     chomp;
>     push @scores, $_;
>     $dis->add_score($_);
> }
>
> if ( $dis->fit_evd() ) {
>     foreach $score (@scores) {
>     $evalue = $dis->evalue($score);
>     print "S: $score   E: $evalue\n";
>     }
> }
> else {
>     warn("Cannot fit!");
> }
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l
>

--
Jason Stajich
Duke University
http://www.duke.edu/~jes12




More information about the Bioperl-l mailing list