BEST

Software Category
Version
3.1.2

BEST

BEST, which is an acronym for Bayesian Estimation of Species Trees, is a bioinformatics program designed to analyze phylogenetic trees built out of multilocus molecular sequence data and estimate their joint posterior distributions in a Bayesian framework. This software is designed for use with the MrBayes phylogenetic tree building program. The program can account for deep coalescence but is not designed to address horizontal gene transfer or gene duplication issues among others.

Using BEST on RCC Resources

BEST requires the intel and openmpi modules to be run on HPC login nodes and on Spear. To load these modules, simply type module load intel openmpi into the command line.

Interactively Running BEST

BEST can be run in a couple of different modes. In order to start an interactive session in BEST, all you need to do is load the necessary modules and call BEST from the command line:

module load intel openmpi
best

Non-Interactive Jobs with BEST

If you want to run BEST non-interactively with a genome file, you can run it on the HPC login nodes or on Spear by running the following commands, where EXT is a bioinformatics file format like Nexus (.nex) and FILE is the name of the data file. Example data files are available on the official website.

module load intel openmpi
best FILE.EXT

To run BEST non-interactively in parallel with a genome file, you can run it on the HPC compute nodes or on Spear by writing a Slurm script. The script must be saved as a file with the .sh suffix; we use the name BESTJOB.sh for the example below. The Slurm script must contain the commands in the first box below, and then must be run on the command line using the command in the second box below. In the script, EXT is a bioinformatics file format like Nexus (.nex) and FILE is the name of the data file (note that example data files are available on the official website).

#!/bin/bash
#SBATCH -J BestJob   # Can rename the job whatever you want
#SBATCH -n 16        # Can use more or less processes if you want
#SBATCH -p genacc_q  # Can use any queues you have access to
#SBATCH -t 1:00:00   # Can request more time in HH:MM:SS format if needed

module load intel openmpi
srun best FILE.EXT
sbatch BESTJOB.sh

Alternatively, you can run BEST in parallel using the Slurm scheduler on multiple nodes. This can be done, for example, with the above Slurm script. Again, you can find example data files on the official website.