HyPhy message board
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl
Methodology Questions >> How to >> Multiple data sets
http://www.hyphy.org/cgi-bin/hyphy_forums/YaBB.pl?num=1371137785

Message started by Nick1979 on Jun 13th, 2013 at 8:36am

Title: Multiple data sets
Post by Nick1979 on Jun 13th, 2013 at 8:36am
Hi,

I am now using the hyphy windows version and I was wondering how to apply the same models and tests to many sets of alignments.


Nicholas

Title: Re: Multiple data sets
Post by Sergei on Jun 13th, 2013 at 2:31pm
Hi Nicolas,

Generally, you would want to write a wrapper to have HyPhy execute the same analysis on different inputs. Assuming that you have a standard analysis (or wrote one yourself), you could do something along the lines of Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login Login

Sergei

Title: Re: Likelihood function constraints
Post by Nick1979 on Jun 17th, 2013 at 9:16am
Hi Sergei,

I have two aligned sequences and i want to test whether the three codon sites evolve at equal rates.

So far I have the code below. Now I want to put a constraint in which sites 2 and 3 have the same branch length as site 1 and then compare the two optimized likelihood functions with an LRT. I will greatly appreciate if you help me with the coding ?  Thank you




DataSet myData = ReadDataFile ("C:/Users/bio/Desktop/example");


/* Create 3 filters: one for each of the
   three codon positions */

DataSetFilter myFilter1 = CreateFilter (myData,1,"<100>","0,1,3");
DataSetFilter myFilter2 = CreateFilter (myData,1,"<010>","0,1,3");
DataSetFilter myFilter3 = CreateFilter (myData,1,"<001>","0,1,3");


HarvestFrequencies (obsFreqs1, myFilter1, 1, 1, 1);
HarvestFrequencies (obsFreqs2, myFilter2, 1, 1, 1);
HarvestFrequencies (obsFreqs3, myFilter3, 1, 1, 1);


/* Define an HKY85 model using each of the 4 frequency sets */
/* Let R be the ts/tv ratio, which we will eventually allow to be shared by all three of the positions */
global R;

HKY85RateMatrix = {{*,b,R*b,b}{b,*,b,R*b}{R*b,b,*,b}{b,R*b,b,*}};
Model HKY851 = (HKY85RateMatrix, obsFreqs1);
Tree myTree1 = (a,b);
Model HKY852 = (HKY85RateMatrix, obsFreqs2);
Tree myTree2 = (a,b);
Model HKY853 = (HKY85RateMatrix, obsFreqs3);
Tree myTree3 = (a,b);



/* Analyze data using 3 partitions: each partition has separate rates and base frequencies, but the ts/tv ratio (R) is shared by all three. Also, each branch has the same ts/tv ratio. */


LikelihoodFunction theSplitLikFun = (myFilter1,myTree1,myFilter2,myTree2,myFilter3,myTree3);


Optimize (paramValues, theSplitLikFun);
lnLik1 = paramValues[1][0];
npar1 = paramValues[1][1]+9;

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