Automating intron discovery in EST-derived gene models

[maizi]

I was trying to do a very similar thing a few years back, and initially I tried mapping my ESTs to the genomic data use est2genome and then trying to interpret the gene structure. In short, I ended up using Scipio (not the web-based version) and it worked very well. Just pick a closely related species' genome if your species of interest has not been sequenced yet.

Hi,I am using scipio now,according to the manual, I need to use perl script scipiogff2gff.pl to convert the file format to .gff, but I can't find it...
can you help me?
Many thanks

[SES]

I was trying to do a very similar thing a few years back, and initially I tried mapping my ESTs to the genomic data use est2genome and then trying to interpret the gene structure. In short, I ended up using Scipio (not the web-based version) and it worked very well. Just pick a closely related species' genome if your species of interest has not been sequenced yet.

[DFJ111]

So, yes, I have a large number of gene models that are derived from EST data. That is, the coding sequence is correct but any introns will be missing. I also have a bunch of 454 genomic sequence (enriched for the particular genes) which will enable me to find the introns.

Obviously I can just map the raw 454 genomic sequence to the gene models to find where introns are, and I have done this. However I don't really want to do it manually for each gene model.

I was thinking to derive some sort of consensus sequence from the genomic sequence mapped onto the (reference sequence) gene model, with a large tolerance for introducing gaps in the gene model (which would be introns). At a bit of a loss as to how to do this. Any suggestions on how to automate this process?