Assembling ESTs and illumina reads

Question

[Joseph Hughes]

I am trying to improve the annotations in a region of the Sheep genome and have compiled at set of 60 Sanger sequenced ESTs from sheep which I would like to assemble with a set of 2,863,891 sheep Illumina reads from a transcriptomic study. These are the reads that mapped to a syntenic region in the cow genome (this being a much better assembly and well annotated). I want to combine these two different sets to obtain the best possible set of transcripts for this region of the sheep genome. I've looked at Trinity but have not seen an option for incorporating ESTs.

I would be grateful for any suggestions.

[h.mon]

Maybe you could assemble the Illumina reads and map them (or map the reads directly) and map the ESTs separately, then check for overlap.

As an alternative, you could use iAssembler to join an Illumina Trinity assembly with the ESTs, but with so few ESTs, this would be too much time-consuming for a small return. I feel mapping and getting the overlaps will be faster and simpler.

edit: with that few ESTs, you can even check manually the overlaps with any genome browser.

[Joseph Hughes]

I'm not having much luck running iAssembler. At the moment, I'm thinking of doing a de-novo transcriptome assembly with Trinity and then using CPA3 to merge my ESTs with the trinity-assembled transcripts.