[Genome] Blat question

[Wes Barris]

Hi,

I have been using blat a lot for the past few years.  I have just encountered
something that I have not seen before.  I am blatting a sequence against
Btau3 BTA18 like this:

blat BTA18.fa in.fa out.psl -minIdentity=90 -t=dna -q=dna -out=psl -noHead -ooc=11.ooc -mask=lower -qMask=lower

The psl output produced looks looks like this:

889     64      0       0       2       -110    2       633138  -       oacn14 843      0       843     BTA18   62891490        56102099        56736190       4133,87,463,270, 0,136,223,573, 
56102099,56102232,56102616,56735920,

When I convert this to a side-by-side alignment, it produces HSPs that overlap.
In the attached file, notice that the first HSP ends at query coordinate 270
and that the second HSP starts at query coordinate 158.  Both HSPs are using
an overlapping portion of the query sequence.  The target coords are from
different areas of BTA18.

In all of the blatting I have done, I have never seen this happen.  Is this
simply a very unique case or is something wrong?

I have attached both the in.fa file and the out.sbs file.
-- 
Wes Barris <wes.barris at csiro.au>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: in.fa
Url: http://www.soe.ucsc.edu/pipermail/genome/attachments/20061120/6ef4109c/in.bat
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: out.sbs
Url: http://www.soe.ucsc.edu/pipermail/genome/attachments/20061120/6ef4109c/out.bat