Difference between revisions of "MrBayes"

From UFRC
Jump to navigation Jump to search
m (Text replacement - "#uppercase" to "uc")
(3 intermediate revisions by 3 users not shown)
Line 7: Line 7:
 
|{{#vardefine:exe|}} <!--Present manual instructions for running the software -->
 
|{{#vardefine:exe|}} <!--Present manual instructions for running the software -->
 
|{{#vardefine:conf|}} <!--Enable config wiki page link - {{#vardefine:conf|1}} = ON/conf|}} = OFF-->
 
|{{#vardefine:conf|}} <!--Enable config wiki page link - {{#vardefine:conf|1}} = ON/conf|}} = OFF-->
|{{#vardefine:pbs|1}} <!--Enable PBS script wiki page link-->
+
|{{#vardefine:pbs|}} <!--Enable PBS script wiki page link-->
 
|{{#vardefine:policy|}} <!--Enable policy section -->
 
|{{#vardefine:policy|}} <!--Enable policy section -->
 
|{{#vardefine:testing|}} <!--Enable performance testing/profiling section -->
 
|{{#vardefine:testing|}} <!--Enable performance testing/profiling section -->
Line 20: Line 20:
 
MrBayes is a program for Bayesian inference and model choice across a wide range of phylogenetic and evolutionary models. MrBayes uses Markov chain Monte Carlo (MCMC) methods to estimate the posterior distribution of model parameters.
 
MrBayes is a program for Bayesian inference and model choice across a wide range of phylogenetic and evolutionary models. MrBayes uses Markov chain Monte Carlo (MCMC) methods to estimate the posterior distribution of model parameters.
  
In addition to the official MrBayes 3.2.1 distribution, a modified version that uses the Gamma-Dirichlet and Inverse Gamma-Dirichlet priors for branch lengths (Rannala et al. in press; Zhang et al. 2012). This also includes the two exponential priors on internal and external branch lengths described by Yang & Rannala (2005) and Yang (2007). Please see [http://abacus.gene.ucl.ac.uk/software.html Ziheng Yang's site] for more information and citations. The modified version is only installed for openmpi and is called mb_Dirichlet. To run this version, simply replace mb with mb_Dirichlet in the examples below.
+
In addition to the official MrBayes distribution, a modified version that uses the Gamma-Dirichlet and Inverse Gamma-Dirichlet priors for branch lengths (Rannala et al. in press; Zhang et al. 2012). This also includes the two exponential priors on internal and external branch lengths described by Yang & Rannala (2005) and Yang (2007). Please see [http://abacus.gene.ucl.ac.uk/software.html Ziheng Yang's site] for more information and citations. The modified version is only installed for openmpi and is called mb_Dirichlet. To run this version, simply replace mb with mb_Dirichlet in the examples below.
  
 
<!--Modules-->
 
<!--Modules-->
Line 35: Line 35:
 
*{{#var:app}}
 
*{{#var:app}}
 
==System Variables==
 
==System Variables==
* HPC_{{#uppercase:{{#var:app}}}}_DIR - installation directory
+
* HPC_{{uc:{{#var:app}}}}_DIR - installation directory
 
<!--Additional-->
 
<!--Additional-->
 
{{#if: {{#var: exe}}|==Additional Information==
 
{{#if: {{#var: exe}}|==Additional Information==
Line 44: Line 44:
 
For MPI jobs: Remember that the total number of cores cannot be greater than the total number of chains MrBayes is using. For example, if you have 2 runs, each with 4 chains, the maximum number of cores to request is 2*4=8 cores. As long as the number of cores is under 12, it is best to keep nodes=1 (nodes = physical machines) as that will keep all communication within a single node. If you need more than 12 cores, you will probably want to add :infiniband to the request. (e.g. nodes=4:ppn=8:infiniband).
 
For MPI jobs: Remember that the total number of cores cannot be greater than the total number of chains MrBayes is using. For example, if you have 2 runs, each with 4 chains, the maximum number of cores to request is 2*4=8 cores. As long as the number of cores is under 12, it is best to keep nodes=1 (nodes = physical machines) as that will keep all communication within a single node. If you need more than 12 cores, you will probably want to add :infiniband to the request. (e.g. nodes=4:ppn=8:infiniband).
  
 +
<!--
 
{{#fileAnchor: mrbayes.MPI.pbs}}
 
{{#fileAnchor: mrbayes.MPI.pbs}}
 
Download raw source of the [{{#fileLink: mrbayes.MPI.pbs}} mrbayes.MPI.pbs]
 
Download raw source of the [{{#fileLink: mrbayes.MPI.pbs}} mrbayes.MPI.pbs]
Line 102: Line 103:
 
mb my_nexus.nex
 
mb my_nexus.nex
 
</source>
 
</source>
 
+
-->
 
{{#if: {{#var: policy}}|==Usage Policy==
 
{{#if: {{#var: policy}}|==Usage Policy==
 
WRITE USAGE POLICY HERE (perhaps templates for a couple of main licensing schemes can be used)|}}
 
WRITE USAGE POLICY HERE (perhaps templates for a couple of main licensing schemes can be used)|}}

Revision as of 21:23, 6 December 2019

Description

mrbayes website  

MrBayes is a program for Bayesian inference and model choice across a wide range of phylogenetic and evolutionary models. MrBayes uses Markov chain Monte Carlo (MCMC) methods to estimate the posterior distribution of model parameters.

In addition to the official MrBayes distribution, a modified version that uses the Gamma-Dirichlet and Inverse Gamma-Dirichlet priors for branch lengths (Rannala et al. in press; Zhang et al. 2012). This also includes the two exponential priors on internal and external branch lengths described by Yang & Rannala (2005) and Yang (2007). Please see Ziheng Yang's site for more information and citations. The modified version is only installed for openmpi and is called mb_Dirichlet. To run this version, simply replace mb with mb_Dirichlet in the examples below.

Required Modules

modules documentation

Serial

  • mrbayes

Parallel (OpenMP)

  • intel
  • mrbayes

Parallel (OpenMPI)

  • intel
  • openmpi
  • mrbayes

System Variables

  • HPC_MRBAYES_DIR - installation directory


For MPI jobs: Remember that the total number of cores cannot be greater than the total number of chains MrBayes is using. For example, if you have 2 runs, each with 4 chains, the maximum number of cores to request is 2*4=8 cores. As long as the number of cores is under 12, it is best to keep nodes=1 (nodes = physical machines) as that will keep all communication within a single node. If you need more than 12 cores, you will probably want to add :infiniband to the request. (e.g. nodes=4:ppn=8:infiniband).