I am looking at a couple of unpaired prokaryotic RNA-Seq datasets. These data have been obtained from bacteria (cultures likely axenic, not guaranteed). My aim is to figure out whether my genes of interest are transcribed in operons.

My first attempt was to use BBDuk for quality control, then rnaSpADES for assembly. It worked a little bit...but I had so many chimeras the results were meaningless.

I thought maybe reference-guided approach would be be better. So I used Rockhopper (https://cs.wellesley.edu/~btjaden/Rockhopper/index.html), which is meant for exactly this - operon prediction from RNA-seq. However, the operon predictions were...not quite as expected? For example, let's say I was expecting a 10-gene operon, with the genes in the same orientation being A, B, C, D, E, F, G, H, I, and J. Results: A, B, H are not listed; C-G form an operon, I-J another. Of course, it could be that genes A, B and H are really expressed independently, but that is doubtful based on the distribution of intergenic regions (for example, there is practically no space between G and H) and function.

Would you be able to suggest an alternative for operon prediction from RNA-seq (reference genome available), or otherwise have an idea what went wrong with the Rockhopper results? Thank you for your time!

Edit: enter image description here I've had a look at Rockhopper results with IGV in more detail. Still can't understand it. It looks like there's enough expression, even in intergenic regions, but for some reason, only one operon automatically defined.

  • $\begingroup$ Maybe take look at the alignment on which rockhopper based the operons. I assume it determines operons by detecting transcripts that span 2 genes? If there are no reads in your alignment that span genes G and I Rockhopper has no evidence that they are in an operon together. $\endgroup$
    – Pallie
    Commented Jun 27 at 13:39
  • $\begingroup$ @Pallie Thank you! That's a good point. I've had a look with IGV, but this doesn't seem to be the case? Confusing. $\endgroup$
    – Laura
    Commented Jul 8 at 10:16


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.