[Bioperl-l] help with BLAST

Remo Sanges sanges at biogem.it
Mon Nov 3 07:16:01 EST 2003


Hi Becca,

I think it is wrong to use blast in order to search for particular element.
Blast algorithm is a local alignment so it is normal to have multiple HSP,
particularly if you search for repetitive elements.
In order to search for LINE you can use repat masker
Its output look like this:

   SW  perc perc perc  query       position in query    matching       repeat        position in  repeat
score  div. del. ins.  sequence    begin  end (left)   repeat         class/family   begin  end (left)  ID

  226  23.7  0.0  1.7  162            96  154  (372) +  L1ME           LINE/L1         5544 5601  (569)   1
  483  11.1  1.2  0.0  162           441  521    (5) +  B1_MM          SINE/Alu          31  112   (35)   2
 1869   7.5  0.7  0.7  175            11  291  (286) +  RLTR4_MM       LTR/ERV1         462  742    (0)   3
.............................................

And in BioPerl there is a module to parse this output:
Bio::Tools::RepeatMasker


Remo
_____________________________________
Remo Sanges - Ph.D. Student
BioGeM - IGB
Gene Expression & Sequencing Core Lab
Via Pietro Castellino 111
80131 Naples - Italy
Tel:+390816132303 - Fax:+390816132262
sanges at biogem.it - sanges at iigb.na.cnr.it
_____________________________________

----- Original Message -----
From: "Crook, R." <rc91 at leicester.ac.uk>
To: <bioperl-l at portal.open-bio.org>
Sent: Monday, November 03, 2003 12:14 PM
Subject: [Bioperl-l] help with BLAST


> Hi,
> I'm a newbie to this site so please be gentle!!
> I'm using blastn and remoteblast to find human LINEs and its not working as I'd expect.  I'm using the stardard code
to get the acc and start of each match but the matches aren't as I expect.  They have been fragmented so even when the
subject matches with itself it divides into three parts.  I could write a long complex prog to find the real start but I
think this would be difficult as the many of the sequences will have two LINEs in them.  Does anyone know a way round
the matching.
>
> I hope thats clear to you.
>
> Becca
> Leicester Uni.
>
> _______________________________________________
> Bioperl-l mailing list
> Bioperl-l at portal.open-bio.org
> http://portal.open-bio.org/mailman/listinfo/bioperl-l




More information about the Bioperl-l mailing list