I have multiple RNASeq datasets from one plant species which does not have an assembled reference genome. But I have access to three whole genome re-sequencing(DNASeq) datasets from individuals in the same species. Any suggestions on how to get the best count for RNASeq in this scenario would be very helpful.
Thanks
2 answers
To quantify gene expression you will need to create a transcripts. Assemble both your DNA and RNA seq data by pooling all data that you have (use the right tool for each type of data).
Now validate your transcripts against the genome (this is not really necessary but helps). Use the validated transcripts and compute the sample specific coverages for these.
I'd take a de novo and read-mapping approach to arrive at a usable transcriptome. Although your species seem closely related, transcriptome read-mapping assemblies can be convoluted and leave out useful information. So be aware of the existence of isoform/spliced variants. Try the Tuxedo packages TopHat and Cufflinks.
Log in to answer this question.