Amber12

Description

From the Amber homepage: “Amber” refers to two things: a set of molecular mechanical force fields for the simulation of biomolecules (which are in the public domain, and are used in a variety of simulation programs); and a package of molecular simulation programs which includes source code and demos.

The Amber software suite is divided into two parts: AmberTools13, a collection of freely available programs mostly under the GPL license, and Amber12, which is centered around the sander and pmemd simulation programs, and which continues to be licensed as before, under a more restrictive license.

Version

• Amber12 u21 & AmberTools13 u26

Authorized Users

• Amber12 is available only to USF research groups that already have an existing Amber license.

Platform

• CIRCE cluster

Running Amber12 on CIRCE

The Amber12 manual is essential to understanding the application and making the most of it. The guide and this page should be all you need to get started with your simulations.

• Amber12 Manual: /apps/amber12/doc/Amber12.pdf

• AmberTools13 Manual: /apps/amber12/doc/AmberTools13.pdf

Modules

Amber12 requires the following module file to run:

• apps/amber/12
  • See Modules for more information.

• NOTE: This module MUST be loaded in order for the application and it’s tools to run. Use:

[user@host ~]$ module load apps/amber/12

Submitting a Job

• NOTE: Make sure to run your job from your $WORK directory.

You must create a submit script (for example, “amber_test1.sh”) like the following:

#!/bin/bash
#
#SBATCH --nodes=2 --ntasks-per-node=4 
#SBATCH --job-name=amber_test1
#SBATCH --output=amber.%j.output
#SBATCH --time=01:00:00

#### Slurm task using 2 Nodes with 4 Processors per Node for Amber test to run for 1 hour.

module purge
module load apps/amber/12

mpirun -np $SLURM_NTASKS -ppn $SLURM_NTASKS_PER_NODE sander.MPI -ng <# groups> -groupfile 

The above script can be copied into your job directory (the folder with your input files) and modified so that you can submit batch processes to the queue. Scripts are provided as examples only. You will need to make modifications before the script can be used. Your executables, tools, and options will vary. For help on submitting jobs to the queue, see our SLURM User’s Guide.

Next, make sure that you are in your job’s directory, and run the sbatch command to submit the job:

cd $WORK/my/job/directory
sbatch ./amber_test1.sh

• You can view the status of your job with the “squeue -u <username>” command
  • Please note that you should alter the number of nodes, and number of processors per node requested to appropriately balance your CPU and node count. For example, for a job requiring a CPU count of 20, the following lines would properly balance your CPU/Node count via SLURM:

-N 2 --ntasks-per-node=10 (2 nodes, 10 cpus per node)
-N 4 --ntasks-per-node=5 (4 nodes, 5 cpus per node)
-N 5 --ntasks-per-node=4 (5 nodes, 4 cpus per node)

Benchmarks and Tests

The following contain input parameters for known Amber12 tests:

• /apps/amber12/benchmarks
• General Description of the Programs
• Online Tutorials
• Downloadable Benchmark Suites

Additional Documentation

• /apps/amber12/doc
• http://ambermd.org/doc12/Amber12.pdf
• http://ambermd.org/doc12/AmberTools13.pdf

More Job Information

See the following for more detailed job submission information:

• SLURM User’s Guide
• Scheduling and Dispatch Policies
• Advanced Submit Techniques

Reporting Bugs

Report bugs with Amber12 to the IT Help Desk: rc-help@usf.edu