mapping ESTs to genome

[Jose Blanca]

Hi:

I have some 454 and sanger reads that I want to map to a genome. I've tried with gmap. gmap find the introns, but tends to do a lot of small misalignments inside the exons that end up as false positive SNPs.
I was wandering about possible alternatives. I've read the tophat page, but I don't know if it would be ok to map long reads (sanger and 454) with tophat. tophat is based on bowtie in the bowtie website the do not recommend to use it with long reads.
Any advice on this matter.
Best regards,

Jose Blanca

[natstreet]

I normally use exonerate for this. Some alternative would include BLAT, sim4db and genome threader.

[Jose Blanca]

Thanks for the tips, I'll look into them. I didin't know about sim4db and genome threader. sim4db seems interesting, the only problem is that it doesn't support sam output.
exonerate is nice, but it is not possible to use it to align ESTs to a whole genome.

[natstreet]

I frequently use exonerate to align ESTs to a genome using exonerate server. For example

fasta2esd genome.fasta genome.esd
esd2esi genome.esd genome.esi --memorylimit 2048

exonerate-server genome.esi --port 12886 &

exonerate --model est2genome ESTs.fasta  localhost:12886 --percent 70 --score 100 --showvulgar yes --softmaskquery no --softmasktarget yes --minintron 20 --maxintron 6000 --ryo ">%qi length=%ql alnlen=%qal\n>%ti length=%tl alnlen=%tal\n" --showalignment no  --showtargetgff yes --geneseed 250  > $file.exonerate

Make sure to set percent and maxintron to something sensible for your situation. You can later parse the exonerate output to get GFF. I guess it would also be possible to write a parser to get SAM formatted output but this I don't have.

The main limitation I find with exonerate is that it's not multi threaded so you have to write your own wrapper to split a large input set of EST and distribute them across threads / nodes.