Entering edit mode
2.7 years ago
changxu.fan
▴
70
Dear community,
I'm looking for a package that tests for positive selection for a given branch. Importantly, non-coding regions must be supported. PAML seems to only have dN/dS, which cannot be used for non-coding regions. data monkey appears to only support dN/dS as well. I've seen a few papers using various methods to detect positive selection in non-coding regions, but couldn't find any software packages that implement them.
Thank you so much!
You can use HyPhy with an integrated batch file (.bf) that could be obtained from here, with this integration you can perform branch-specific analysis of selection acting on non-coding regions. For this, you will need to install hyphy/hyphympi on your local machine and use the .bf file that is given in the link above.
Thank you so much for the answer! While I have you here, can I also ask for a tool that detects purifying selection among a few branches?
Thanks again!!
I think there are estimates like the GERP score which might be of interest for you in that case, although I have personally never used GERP. All these tools centrally rely on the changes in the rate of substitutions. Even with the integration of HyPhy that I described in my previous comment, there is the facility of performing a site-specific analysis of selection on your branch of interest using Bayes Estimates. Here, the estimates will give you site-specific probabilities of it being under neutral/positive/purifying selection. If you are interested in reading about this a bit more, I recently co-authored a review on tools and studies performing analysis of selection on the non-coding regions that you can find here. You may find tools/studies that might be of interest to you. It goes without saying that there is a conflict of interest since I am one of the authors of the said manuscript.