Modeltest-ng and MrBayes

Question

[Francisco Pina Martins]

Hi everyone,

I have recently started using modeltest-ng for my phylogenetic analyses model selection. The program conveniently outputs the model (and even the command line) to use it with popular phylogenetics software like raxml, iqtree, paup, or raxml-ng. This is the easy part. I am, however planing on using it with MrBayes. The thing is, as far as I can tell, modeltest-ng does not produce a MrBayes block to use in a #NEXUS file, or the parameters that will effectively translate the best model to MrBayes. I know this can be done manually, And interpret modeltest-ng's output as lset or prset commands for the #NEXUS block. However, I'm trying to automate my analyses, and as such I would like, as much as possible to find a way to perform this "translation" automatically. Is there any tool that will allow me to do this? Or is it rather a script I should write (and that can maybe be pull requested for inclusion with modeltest-ng)? Thank you in advance for any insights.

Answer 1

[Mensur Dlakic]

It would have been helpful to state whether you are working with DNA or proteins.

I don't think there is a tool for automatic translation from the ModelTest output to MrBayes. Also keep in mind that MrBayes doesn't have all the models that are available out there.

This may help you get going, and I suggest you also look at the links on the right side of this page.

• Best Model

• ##GTR lset applyto=() nst=6 # GTR lset applyto=() nst=6 rates=propinv # GTR + I lset applyto=() nst=6 rates=gamma # GTR + gamma lset applyto=() nst=6 rates=invgamma # GTR + I + gamma

  ##SYM

  lset applyto=() nst=6                           # SYM
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=6 rates=propinv             # SYM + I
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=6 rates=gamma               # SYM + gamma
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=6 rates=invgamma            # SYM + I + gamma
  prset applyto=() statefreqpr=fixed(equal)
  

  ##HKY

  lset applyto=() nst=2                           # HKY
  lset applyto=() nst=2 rates=propinv             # HKY + I
  lset applyto=() nst=2 rates=gamma               # HKY + gamma
  lset applyto=() nst=2 rates=invgamma            # HKY + I + gamma
  

  ##K2P

  lset applyto=() nst=2                           # K2P
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=2 rates=propinv             # K2P + I
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=2 rates=gamma               # K2P + gamma
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=2 rates=invgamma            # K2P + I + gamma
  prset applyto=() statefreqpr=fixed(equal)
  

  ##F81

  lset applyto=() nst=1                           # F81
  lset applyto=() nst=1 rates=propinv             # F81 + I
  lset applyto=() nst=1 rates=gamma               # F81 + gamma
  lset applyto=() nst=1 rates=invgamma            # F81 + I + gamma
  

  Jukes Cantor

  lset applyto=() nst=1                           # JC
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=1 rates=propinv             # JC + I
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=1 rates=gamma               # JC + gamma
  prset applyto=() statefreqpr=fixed(equal)
  
  lset applyto=() nst=1 rates=incgamma            # JC + I + gamma
  prset applyto=() statefreqpr=fixed(equal)
  

  view raw gistfile1.md hosted with ❤ by GitHub

[Francisco Pina Martins]

Thank you for your reply Mensur Dlakic . I have only seen it now (somehow I received no email notification). This is probably worth scripting somehow. I will try to wrap something up with it, and if I ever do I will post back here.

Best,

Francisco