HyPhy message board
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl
Methodology Questions >> How to >> Likelihood function constraints
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1371485989

Message started by Nick1979 on Jun 17th, 2013 at 9:19am

Title: Likelihood function constraints
Post by Nick1979 on Jun 17th, 2013 at 9:19am
Hi Sergei,

I am new at this and would greatly appreciate if you check my code. I have two aligned sequences and i want to test whether the three codon sites evolve at equal rates.

I have attached my code


  Thank you













http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?action=downloadfile;file=prog.bf (2 KB | 8 )

Title: Re: Likelihood function constraints
Post by Sergei on Jun 24th, 2013 at 8:45pm
Hi Nick,

Your code looks good, apart from the fact that comments in HyPhy are C-style ( '//' or '/* */' ), and not Perl (#) style.
One minor point is that in your code, the branch lengths are still slightly different even AFTER you constrain 'b', because the base frequencies are different among partitions.

You could uses pooled base frequencies among all partitions (i.e. the same HKY85 model) which doesn't have this issue; if you need to maintain different sets of frequencies, some transformations of 'b' are required (see the attached file).

Sergei


http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?action=downloadfile;file=separate_freqs.bf (3 KB | 4 )

HyPhy message board » Powered by YaBB 2.5.2!
YaBB Forum Software © 2000-2024. All Rights Reserved.