Mr. Bayes Protein Models OR How to implement prottest output in mrbayes?

Question

[kanwalashraf07]

Hello, I am using Mr. Bayes for my amino acid data set, The model predicted by prottest is rtRev + G + I (rtRev+G+I). I can specify rtRev on mr bayes by using prset aamodelpr=fixed <...> , However, I am not sure how can I implement G + I. Any help will be much appreciated... :)))

---

Mr. Bayes modelling cheat sheet

Edit (MD): Is there a simple list or table overview/command reference how to convert the winning model output from ProtTest to a model in Mr. Bayes using commands lset and prset? Background: I ran my multiple alignment file of through prottest to calculate the optimal amino-acid / nucleotide substitution matrix for my protein or DNA sequence.

Now I want to implement the resulting model in Mr. Bayes to calculate a phylogenetic tree. My model is rtRev+G+I (gamma distributed plus invariant sites).

[Michael]

I have rewritten the question putting in the maximum number of key words for searching.

Answer 1

[arnstrm]

Here is the list for your reference:

I    : lset rates=propinv;
G    : lset rates=gamma;
F    : prset statefreqpr=fixed(empirical);
I + G    : lset rates=invgamma;
G + F    : prset statefreqpr=fixed(empirical); lset rates=gamma; 
I + F    : prset statefreqpr=fixed(empirical); lset rates=propinv;
I + G + F    : prset statefreqpr=fixed(empirical); lset rates=invgamma;

I also add these to my mb code block

mcmcp ngen=1000000 nchains=4 printfreq=100 samplefreq=100 burnin=200; # Number of cycles, sampling/printing frequency and burnin
mcmc; # mcmc analysis
sump; # summarized parameters
sumt; # summarized trees
quit

EDIT: fixed I + G typo

[Whetting]

Are you sure? I believe I+G is invgamma...?! (http://bodegaphylo.wikispot.org/MrBayes_Tutorial)

[arnstrm]

sorry, my bad! it was a bad copy/paste job. Fixed it now, thanks.

Answer 2

[Whetting]

would you not just "lset rates=invgamma"?
www.molecularevolution.org/molevolfiles/mrbayes/Appendix_Fig2.pdf‎ )