[Genome] Blat question

[Wes Barris]

Hi,

I have found a number of cases where stand-alone blat is returning
curious results.  If I blat a masked version of NCBI sequence EE753431
against the latest (2006) version of the bovine genome on the UCSC
web site, the best hit is against BTA9.

If I blat this same sequence using the stand-alone software (v34), the best
hit is against BTA7 and the psl file contains overlapping HSPs.  Here
is the command line that I used and the psl result:

blat db.fa query.fa -t=dna -q=dna -qMask=lower -mask=lower -minIdentity=94 junk.psl

402     23      0       0       1       -93     2       986136  +       EE753431    436     2       334     BTA7.nib:86197000-87184000 987000 17       986578  3       84,192,149,     2,86,185, 
17,460266,986429,

Notice the negative number in column 6.  Is this expected?

I have attached a small portion of BTA9 and the query.
-- 
Wes Barris <wes.barris at csiro.au>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: query.fa
Url: http://www.soe.ucsc.edu/pipermail/genome/attachments/20080125/7354ab43/attachment-0001.ksh