Rui Borges

Hello and welcome to my website.

Home
Education
Professional experience
Publications
Teaching
Software
PoMo-cod

Assignment 1

Fitting models of nucleotide sequence evolution

In this assignment, you will determine the best fitting model of evolution for the multiple sequence alignment assigned to you by employing a hierarchical likelihood ratio test (hLRT).

  1. Test the following models JC, K80, F81, HKY, and GTR, and indicate the best fitting model. Please provide a table including the model likelihoods, number of parameters, model comparisons and corresponding degrees of freedom, the resulting LRT values, and critical values that you used to assess statistical significance. Perform the hLRT using a significance level of 0.05.

  2. Based on the best fitting model obtained previously, do your sequences have a heterogeneous base composition? Is one exchangeability rate enough to describe the evolution of your sequences? Justify your answers.

  3. Now that you know the best fitting model of evolution for your sequences, you will now test for the need for site heterogeneity. Again you will need to perform an LRT test; perform this LRT using a significance level of 0.05. To include the gamma rate in your analyses, open your baseml.ctl file, set the best fitting model, and change the options of fix_alpha and ncatG accordingly. You can use four gamma categories, which is typically a good approximation for the gamma distribution. Is there evidence for site heterogeneity? Based on the estimated value of alpha, is there evidence for strong or weak site rate heterogeneity over sites? Justify all your answers.

Some notes: