I've assembled a transcriptome using trinity in which we are looking for the full sequence of a protein channel. We have found a slight extension of the published sequence to the 3' end but still not the complete sequence. There is a complete sequence in a similar species. I was planning on mapping loosly the reads to the similar sequence to check the reads are there to assemble and then maybe try using bowtie2 in trinity, or SOAPtrans, velvet/Oases? Any ideas on possible routes to assembling one gene of interest from transcriptome data as does not seem to be successful so far.
1 answer
Hi,
you could have a look here:SSPACE, or maybe this L-Assembly
After the initial Assembly step SSPACE tries to enlarge contigs with orphan / leftover reads which sometimes works quite well. L-Assembly, I haven't used it up to now tbh.
planning on mapping loosly the reads to the similar sequence
If the sequence you are mapping to is a (close) relative than this is, imho, valid. Else you might hit some regions just by chance.
best, Phil
Log in to answer this question.
Do you have a genome? Or some set of contigs that can be used for reference-guided assembly? If so you could simply use this to restrict your assembly space.
I have a 3599 bp of the gene but it is predicted to be twice that. Can you use ref sequence in trinity? I've tried mapping to it and taking the reads and unmapped paired reads of those did map and tried trinity just using these reads but no joy.