This is a test version of Biostars. For the public version, visit https://www.biostars.org.
positive selection detection on non-coding dataset using HyPhy

Hi,

I want to detect selection signatures on non-coding dataset via HyPhy version 2.2.0 using the similar strategy as proposed by Haygood et al., 2007 Nat Genet. 39(9):1140-4.

The program provides universal codon option and other options which don't seem to be applicable on my dataset. I am at a loss regarding hyphy usage on non-coding dataset as well positive selection using suitable options e.g., 'yangnielsienbranchsite2005.bf' or else? I am in dire need to get this fixed.

is there anybody who can help me in this regard?

Best wishes

non-coding positive-selection hyphy
• 2,904 views • link
updated by Ram Ram • written by AISHA AISHA

1 answer

I don't know the HyPhy batch language well enough to make a file, but it should be possible to run this kind of analysis. The steps would be

  • For each site you want to create an alignment of your 'focal sequence' and the appropriate null and link each of those to a tree.
  • Fit whatever nucleotide model you like (F81 HKY, REV...) to each alignment given the tre, and store the maximized likelihood.
  • Add a constraint, such that the rate of evolution in your first alignment should be equal to rate in your second, and fit the same nucelotide model under this constraint
  • Use a likelihood ratio test to see whether the unconstrained model fits the data signifcantly better than the "rates equal" model
• 0 views • link
updated by Ram Ram • written by David W David W

I've figured out the possible model to answer my question in HyPhy. However, now i am unable to interpret the output.. which consists of log-likehood value, LRT value and LRT-Pvalue aswell as site specific inferences of selection signatures.

Any help will be highly appreciated!

• 0 views • link
written by AISHA AISHA

Log in to answer this question.

More posts like this