I'm working on a somewhat unusual transcriptome focused on a killer X chromosome. This chromosome has some new genes, with expression levels quite different from their autosomal paralogs.

I've found that Trinity does an ok job of assembling the distinct genes, but when I try to do a gene-level DE analysis, it lumps the distinct genes together. So, I lose important information by doing a gene-level DE analysis and need to use a transcript-level approach.

I've done blastx with all of my transcripts using diamond, and want to extract the isoform from each gene with the best score (bitscore). I'm trying to write a script to parse this, but my scripting skills aren't up to the task. Does anyone know of (or have) a script that can do this?

  • $\begingroup$ Welcome to the site Ryan. Could you post that script so we can help you building these skills? Also it would be helpful if you explain how do you have organized those files/data $\endgroup$ – llrs Mar 19 '19 at 9:11
  • $\begingroup$ Hi Ryan! You mention that you lose important information by doing gene-level DE analysis. If you retain only a single isoform from each gene but still include isoforms from multiple paralogs, these will presumably continue to be lumped together, correct? On the other hand, if you retain only a single isoform from a single paralog, you've thrown away most of the interesting information. Perhaps you can clarify what you hope the result to be. $\endgroup$ – Daniel Standage Mar 19 '19 at 14:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.