[ensembl-dev] BLAT query enquire

[Harpreet Riat]

Hi Duarte,

The indices we use for BLAT are repeat masked and the sequecne you are trying to BLAT falls in a region that gets masked thus not giving you an exact match as you expected.

To verify this, you can run BLASTN on the same sequence with "Filter query sequences using RepeatMasker" option turned off (inside "Filters and masking options"), and you will see the match at the location from where you copied it.

Cheers,
Harpreet


On Mar 3, 2015, at 11:31 AM, Duarte Molha <duartemolha at gmail.com> wrote:

> anyone?
> 
> =========================
>      Duarte Miguel Paulo Molha      
>          http://about.me/duarte         
> =========================
> 
> On 2 March 2015 at 15:45, Duarte Molha <duartemolha at gmail.com> wrote:
> Dear developers
> 
> Please consider this genomic sequence:
> 
> CTTTCTCTTTCTTTCTTTCTTTCTTTTTTCTTTCTTTTCTTTCTCCCTCTCTCTCTCCCT
> 
> 
> 
> I got it from ensembl directly on GRCh37 on coordinates (X:153,795,306-153,795,365)
> 
> 
> 
> Sequence export:
> 
>  
> 
> http://grch37.ensembl.org/Homo_sapiens/Export/Output/Location?db=core;flank3_display=0;flank5_display=0;output=fasta;r=X:153795306-153795365;strand=feature;coding=yes;cdna=yes;peptide=yes;utr3=yes;exon=yes;intron=yes;genomic=unmasked;utr5=yes;_format=Text
> 
> 
> 
> 
> 
> However, if I blat this very same sequence on your own blat servers I get no perfect hit (please see attached text file).
> 
> 
> 
> If, however, I blat it on UCSC blat servers I get 1 perfect matched hit at the expected coordinates:
> 
> BLAT Search Results
> 
>    ACTIONS      QUERY           SCORE START  END QSIZE IDENTITY CHRO STRAND  START    END      SPAN
> ---------------------------------------------------------------------------------------------------
> 
> browser details YourSeq           60     1    60    60 100.0%     X   +  153795306 153795365     60
> 
> Can you explain the discrepancy?
> 
> Best regards
> 
> Duarte
> 
> 
> _______________________________________________
> Dev mailing list    Dev at ensembl.org
> Posting guidelines and subscribe/unsubscribe info: http://lists.ensembl.org/mailman/listinfo/dev
> Ensembl Blog: http://www.ensembl.info/