[Bioperl-l] Problem w/ extracting start and stop of a blast alignment

Hilmar Lapp hilmarl@yahoo.com
Thu, 23 Aug 2001 13:14:11 +0200


Erik,

could you please provide the same information as for a bug report,
or better for tracking if it really is a bug, submit a bug report
through http://www.bioperl.org/Bugs/

I.e., which version of Bioperl, which OS, which Blast parser,
sequence input, full blast output.

thanks, -hilmar
-- 
-----------------------------------------------------------------
Hilmar Lapp                              email: hilmarl@yahoo.com
A-1120 Vienna
-----------------------------------------------------------------

Erik Richly wrote:
> 
> Hi,
> 
> I'm experiencing a problem in extracting the start and stop position of
> a blast alignment.
> 
> I'm parsing the BlastN output using bioperl with the folloing routine:
> 
> foreach $hit ($blastObj->hits) {
>       $id=$hit->name;
>       $exp=$hit->expect;
>       ($qstart,$sstart)=$hit->start;
>       ($qend,$send)=$hit->end;
>       print $hit->name," $exp\n";
>       print "Query: $qstart - $qend   Subj: $sstart - $send\n";
> }
> 
> Now, when I compare the positions I'm getting using the above routine to
> the original blast output it shows that these positions are completely
> wrong - e.g.:
> 
> Output of program:
> ATL8C18794 0.0
> Query: 536 - 39328   Subj: 1 - 3495
> 
> ATL8C20200 0.0
> Query: 15966 - 19593   Subj: 1 - 3608
> 
> Output of BlastN:
> >ATL8C18794
>           Length = 4868
> 
>  Score = 5366 bits (2707), Expect = 0.0
>  Identities = 2777/2800 (99%), Gaps = 2/2800 (0%)
>  Strand = Plus / Plus
> 
> 
> Query: 32585
> gtggtgta-tttaatac-aggtaacgatttcttcaaggagcagaagtatcctgaggccat 32642
>              |||||||| |||||||| ||||| |||||||||||||  ||||||||||||||||
> ||||
> Sbjct: 1
> gtggtgtattttaatacaaggtaccgatttcttcaagaggcagaagtatcctgagcccat 60
> ...
> ....
> .....
> 
> >ATL8C20200
>           Length = 3608
> 
>  Score = 5247 bits (2647), Expect = 0.0
>  Identities = 2786/2834 (98%), Gaps = 6/2834 (0%)
>  Strand = Plus / Minus
> 
> 
> Query: 16766
> ttggcgatggaacgaactactgtaatactgttcgtcttcttcttgctctctgtagcaatg 16825
>              |||||||||||||||||||||||||||||||||||||||||
> ||||||||||||||||||
> Sbjct: 2834
> ttggcgatggaacgaactactgtaatactgttcgtcttctttttgctctctgtagcaatg 2775
> ...
> ....
> .....
> 
> Would anybody of you know why the query start/stop positions are messed
> up?
> 
> Thanks in advance,
> 
> Erik
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l@bioperl.org
> http://bioperl.org/mailman/listinfo/bioperl-l