You can place commonly used options in the top of your job scripts. For example,  if you know that you are always submitting to the 'cpu' queue using the 'S1.1' account, and you want to write output to 'output.log' and errors to 'error.log', then you can add the options like this:

#!/bin/bash -login
#SBATCH -p cpu
#SBATCH -A S1.1
#SBATCH -o output.log
#SBATCH -e error.log

run_my_job

Job scripts are just simple shell scripts that contain the commands that you want to run. For example, if you want to run a single GPU pmemd.cuda job, you could use a file called "run.slurm" that contains the following:

#!/bin/bash -login
#SBATCH -p gpu
#SBATCH -A other_mvz
#SBATCH --gres=gpu:1
#SBATCH -N 1
#SBATCH -t 15-0
#SBATCH --ntasks-per-node=1

module add apps/amber-14

pmemd.cuda -O -i mdin -o mdout

You could then submit this job using just the command

• sbatch run.slurm