Utility scripts to run arbitrary bioinformatics software on a PBS server.
Currently only serial jobs (i.e. jobs that run on a single node exclusively) 
are supported. Two _run scripts are provided: garli_run and mb_run for GARLI 
and MrBayes respectively.
Note: it is necessary that all PBS computing nodes are configured to create 
per-job temporary directory and export it's path as TMPDIR environment variable.