Hi,

I try to implement Wong and Nielsen selection test for non-coding sequences in hyphy. Briefly this method is a modification of Yang and Nielsen test on coding regions.The difference is that new method uses two parititions (coding and non-coding) under two models (codon model and HKY). The HKY model has an additonal parameter zeta that is the rate normalized by the sysnonymous rate of the coding partition.

I am a novice in hyphy and can't figure out how to set up this zeta parameter in hyphy.

any advice?

thanks a lot

Olivier

Duke university
Biology Department

Dear Olivier,

To set up this test one needs to write a custom batch file, which I'll be happy to do since it might be of general interest and will take me very little time to write.

As input in can either take:

1). Two separate sequence files (coding and noncoding) but with the same tree and the same number of sequences

2). A joint NEXUS file where coding and noncoding parts are defined as NEXUS CHARSETS

Which is easier?

Cheers,
Sergei

hi Sergei,

it would be great if you can do it. I tried to modify the Yang & Nielsen test batch file but there are quite a bit of functionalities of the code that I do not understand yet.

to answer your question, two separate sequence files (coding and noncoding) will be the best.
eventually I would like to modify it later in order to use an intron as a neutral proxy as well (using the rate of the intron instead of ds to normalize zeta). this will be certainly relatively straighforward from your code.

thanks a lot,

Olivier

Dear Olivier,

Would you mind e-mailing me (help.at.datamonkey.org) some sample data, so I can debug the analysis before I post it?

Cheers,
Sergei

Dear Olivier,

Here's a barebones (but self-contained) Wong-Nielsen analysis. I can't really test it because the Genetics paper doesn't point to any alignments downloadable that I can try it on:(

Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login

Cheers,
Sergei

Hi!

Just to make sure I understood properly, I will need to download HyPhy and implement the batch file there. There is no way of doing this in the datamokey server, right?

I have a very slow computer  :-[

Hi Nica,

You are correct -- these analyses must be run on your computer. On the positive side, they should not be TOO CPU intensive.

Sergei