Kind of curious. If your version of BLAT is a commandline version, then
-minScore=60 -minIdentity=100
In the above case, my query sequence length is 60 and I am requesting
for genome site hits full-score and full percent-identity.
Following is quick way of parsing the raw PSl results from either
command-line or webblat and reformatting the results to calculate the
score and PI.
************************************************************************
***************************
#include "common.h"
#include "sqlNum.h"
#include "sqlList.h"
#include "localmem.h"
#include "psl.h"
#include "hash.h"
#include "linefile.h"
#include "dnaseq.h"
#include "dystring.h"
#include "fuzzyFind.h"
#include "aliType.h"
#include "binRange.h"
#include "rangeTree.h"
int main(int argc, char *argv[]){
printf("BLAT Psl results parser\n");
char * pslFileName = argv[1];
struct psl *pslList = NULL;
pslList = pslLoadAll(pslFileName);
printf("Filename: %s\n", pslFileName);
struct psl *psl = NULL;
FILE *fp;
fp=fopen("ScoredPslOutput.out", "w");
for (psl = pslList; psl != NULL; psl = psl->next){
char * tName = psl->tName;
int tStart = psl->tStart;
int tEnd = psl->tEnd;
char * strand = psl->strand;
int score = pslScore(psl);
int pslProtein = pslIsProtein(psl);
printf("Psl Protein:%d\t", pslProtein);
int sizeMul = pslIsProtein(psl) ? 3 : 1;
printf("Size Mul:%d\t", sizeMul);
printf("%s\t", psl->qName);
int milliBad = pslCalcMilliBad(psl, TRUE);
printf("MilliBad:%d\t", milliBad);
float pi = 100.0 - (milliBad * 0.1);
printf("%f \n", pi);
fprintf( fp, "%s\t%s\t%s\t%d\t%d\t%d\t%f\n", psl->qName,
tName, strand, tStart, tEnd, score, pi);
}
fclose(fp);
return 0;
}
************************************************************************
**********************************
vinodh
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Galt Barber
Sent: Monday, September 13, 2010 6:07 PM
To: Gregory McInnes
Cc: [email protected]
Subject: Re: [Genome] BLAT
You should use pslReps or pslCDnaFilter to filter BLAT psl results.
Here is the BLAT FAQ.
http://genome.cse.ucsc.edu/FAQ/FAQblat.html
On 09/13/10 15:32, Gregory McInnes wrote:
> Hello,
>
> Is there a way to limit the results BLAT returns to only 100% matches?
This
> would be extremely helpful.
>
> Thank you,
>
> Greg McInnes
> _______________________________________________
> Genome maillist - [email protected]
> https://lists.soe.ucsc.edu/mailman/listinfo/genome
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
_______________________________________________
Genome maillist - [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome
