[Genome] BLAT- commandline and web results differ

[Brooke Rhead]

Hello Jenny,

We recommend that you take out the "-mask=lower" option.  This option is 
usually only used for protein or translated blat searches (not for 
aligning DNA/DNA or RNA/DNA).  The parameters for replicating web-based 
blat using stand-alone, command-line blat are:

blat -stepsize=5 -minScore=0 -minIdentity=0 database.2bit query.fa 
output.psl

If you ARE trying to use translated blat, or if you are still having 
problems using the parameters above, perhaps you can send us a specific 
example of a sequence that is giving unexpected results, and we can try 
to provide further assistance.

--
Brooke Rhead
UCSC Genome Bioinformatics Group


Tan Hui Hui Jenny wrote:
> Hi,
> 
> I have tried to replicate the web BLAT results on the standalone blat
> The options I used were:
> blat -stepSize=5 -minScore=0 -minIdentity=0 -mask=lower   (.....etc
> etc....)
> 
> These are exactly the instructions given at
> http://genome.ucsc.edu/FAQ/FAQblat#blat5.
> 
> However, some sequences that I expected to show 100% hits against some
> human chromosomes did not have a hit at all in the standalone blat
> whereas they appeared in the web BLAT results.
> 
> Can someone shed light on the discrepancy or what did I go wrong? 
> 
> Thanks for ur time in answering this.....
> 
> -jenny 
> 
> _______________________________________________
> Genome maillist  -  Genome at soe.ucsc.edu
> http://www.soe.ucsc.edu/mailman/listinfo/genome