>>> browser details hsa-mir-17        84     1    84    84 100.0%    13 
   +   92002859  92002942     84

This is a long (84bp) perfect match.
It is not difficult to find,
it is not short, it is not a repeat.
It is not at the limits of sensitivity.
There is nothing challenging about it.

Except that it has "U" in the input sequence instead of T.
Please try replacing all the U's with T's in your input query file
and you should get the results you expect.

Recently hgBlat was given the ability to tolerate U's,
but none of the command-line blat utilities do at this time.

--

Here's the answer to your other question about psLayout 4 versus 3:

psLayout version 4 seems to have been invented to support pslx,
based on clues in lib/psl.c.

The pslx format stores the query and target sequences,
in additional fields appended to the record.
(However hgBlat does not show these sequences.)

hgBlat writes "psLayout version 4",
and the cart variables say that it is a pslx file,
however, if one looks, it is really only a .psl
file as there is no extra sequences for query and target.
Perhaps the original implementation was actual .pslx
and then they changed their minds.
Instead the queries gets stored in an additional fasta file,
and the target is pulled in from the reference genome.

Probably, hgBlat output should be changed back to say
psLayout 3.  It's using none of the extra fields of pslx or
psLayout 4.

Finally, when I create -out=pslx,
the header of the .pslx file still says psLayout version 3.

Whatever the difference was supposed to mean,
I think you can safely ignore it.

-Galt

Ar 5/17/2010 12:27 AM, scríobh Mauro Donadello:
> Hi Galt,
> thanks for the answer.
> Via command line this is the input :
>
> gfClient -t=dna -q=rna  $HOST $PORT hg19.2bit $infile $outfile
>
> Via Web these are the selected options:
>
> Genome:Human;
> Assembly:GRCh37/hg19;
> QueryType:DNA;
> Sort Output:Query Score;
> Output Type:PSL;
>
> Thanks again.
>
> Ps. Why the output of gfClient-Server is labeled
> psLayout version 3  instead of psLayout version 4 of hgBlat ?
>
> On May 15, 2010, at 1:32 AM, Galt Barber wrote:
>
>>
>> Hi, Mauro!
>>
>> In order to help, we need some more information:
>>
>> What commandline was used to query with gfClient?
>>
>> What commandline was used to start gfServer?
>>
>> -Galt
>>
>> Mauro Donadello wrote:
>>> Hi, i've this sequence :
>>>> hsa-mir-17 MI0000071
>>> GUCAGAAUAAUGUCAAAGUGCUUACAGUGCAGGUAGUGAUAUGUGCAUCUACUGCAGUGAAGGCACUUGUAGCAUUAUGGUGAC
>>> if I use hgBlat(webBlat) of our mirror , I obtain these results :  
>>> Genome:Human; Assembly:GRCh37/hg19
>>>     ACTIONS      QUERY           SCORE START  END QSIZE IDENTITY CHRO 
>>> STRAND  START    END      SPAN
>>> ---------------------------------------------------------------------------------------------------
>>> browser details hsa-mir-17        84     1    84    84 100.0%    13   +   
>>> 92002859  92002942     84
>>> browser details hsa-mir-17        33    10    56    84  89.2%     X   -  
>>> 133304255 133304300     46
>>> browser details hsa-mir-17        22    43    65    84 100.0%  6_qbl_hap6   
>>> +    2248519   2248544     26
>>> browser details hsa-mir-17        22    43    65    84 100.0%  6_mann_hap4  
>>>  +    2303515   2303540     26
>>> but when I use gfClient there are no results.
>>> gfClient and gfServer are compiled from 2.30 gb ucsc source code. Any idea 
>>> ? Thanks a lot. _______________________________________________
>>> Genome maillist  -  [email protected]
>>> https://lists.soe.ucsc.edu/mailman/listinfo/genome
>
>
>

_______________________________________________
Genome maillist  -  [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to