Merlin on Biowulf
MERLIN uses sparse trees to represent gene flow in pedigrees and is one of the fastest pedigree analysis packages around (Abecasis et al, 2002).
For small numbers of Merlin runs, Helix is the simplest system to use. Merlin on Biowulf would typically be used to run large numbers of simultaneous Merlin jobs with different input files or parameters.Submitting a single Merlin batch job
1. Create a script file. The file will contain the lines similar to the lines below between dotted lines. Modify the command on the last line to use your own input file and the desired parameters:
.....................file /home/username/runMerlin........................ #!/bin/tcsh # This file is runMerlin # #PBS -N merlin #PBS -m be #PBS -k oe date /usr/local/bin/merlin -d c1.dat -m c1.map -p simdata --quiet
2. Submit the script using the 'qsub' command, e.g.
qsub -v -l nodes=1 /home/username/runMerlin
Running a 'swarm' of Merlin jobs
The swarm program is a convenient way to submit large numbers of jobs all at once instead of manually submit them one by one.
1. Create a swarm command file containing a single job on each line, e.g.
................this file is /home/username/merlinjobs.......
/usr/local/bin/merlin -d c1.dat -m c1.map -p simdata --quiet
/usr/local/bin/merlin -d c2.dat -m c2.map -p simdata --quiet
/usr/local/bin/merlin -d c3.dat -m c3.map -p simdata --quiet
/usr/local/bin/merlin -d c4.dat -m c4.map -p simdata --quiet
/usr/local/bin/merlin -d c5.dat -m c5.map -p simdata --quiet
[...]
2. Submit this swarm job by typing:
swarm -f /home/username/merlinjobs
Swarm will create the batch scripts and submit the jobs to the batch system.
Documentation
http://www.sph.umich.edu/csg/abecasis/Merlin/tour/linkage.html
