runningBEAST2onCeres.txt

Below are steps for downloading and running the most up to date version of BEAST2 on the Ceres supercomputing cluster. You will require an .xml file generated by an updated version of BEAUTi and access to an account on Ceres.

1.	From Ceres, Open a new Terminal.
 
2.	Start an interactive job using ‘salloc’
 
3.	Enter the following to download the latest version of beast to your personal home directory. (As of 8/27/24, the latest version is 2.7.7)
 
wget https://github.com/CompEvol/beast2/releases/download/v2.7.7/BEAST.v2.7.7.Linux.x86.tgz
tar zxvf BEAST.v2.7.7.Linux.x86.tgz
 
End the interactive job using ‘scancel JOBID#’ (can find job ID # using ‘squeue -u USERNAME’)
 
 
4.	To run a batch script, navigate to your home folder and run ‘sbatch SCRIPTDIRECTORY/SCRIPTNAME.sh’ where the script of SCRIPTNAME.sh looks something like:

#!/bin/bash
 
#SBATCH --job-name=sat1
 
#SBATCH -N 1
 
#SBATCH -n 12  
 
#SBATCH -t 1:59:00  
 
#SBATCH --mem-per-cpu=15000M   
 
module load java
 
~/beast/bin/beast -threads 12 sat1/sat1_beauti.xml
 
Of note, I’m starting from the home directory so I can directly access beast in the /beast/bin directories and my .xml in the sat1 directory. Thus, when I call on my sbatch, it’s ‘sbatch sat1/SCRIPTNAME.sh’. This is probably not the most sophisticated setup, but it did function!