ABySS is a de novo, parallel, paired-end sequence assembler that is designed for short reads. The parallel version is implemented using MPI and is capable of assembling larger genomes.

References:

• https://genome.cshlp.org/content/27/5/768

• https://github.com/bcgsc/abyss

• Module Name: abyss (see the modules page for more information)
• MPI app; when running abyss on multiple nodes, please follow the syntax in the example below instead of using traditional MPI syntax.

Create a batch input file (e.g. batch.sh). For example:

#!/bin/bash
#SBATCH --job-name="abyss"
#SBATCH --mail-type=BEGIN,END
### sbatch --partition=multinode --ntasks=16 --nodes=2 --time=24:00:00 --mem=60g batch.sh

cd /data/$USER/abyss
module load abyss
`which abyss-pe` np=${SLURM_NTASKS} j=8 k=25 n=10 in='/data/$USER/File_1.fq /data/$USER/File_2.fq' name=OutputPrefix

Submit this job using the Slurm sbatch command.

sbatch --partition=multinode --ntasks=16 --nodes=2 --time=24:00:00 --mem=60g batch.sh

The job runs 16 tasks ( np=${SLURM_NTASKS} ) on 2 nodes with 8 (j=8) cpus each node.

Create a swarmfile (e.g. job.swarm). For example:

cd dir1;abyss-pe np=${SLURM_NTASKS} j=8 k=25 n=10 in='1.fq 2.fq' name=out
cd dir2;abyss-pe np=${SLURM_NTASKS} j=8 k=25 n=10 in='1.fq 2.fq' name=out
cd dir3;abyss-pe np=${SLURM_NTASKS} j=8 k=25 n=10 in='1.fq 2.fq' name=out

Submit this job using the swarm command.

swarm -f job.swarm --module abyss --sbatch "--partition=multinode --ntasks=16 --nodes=2 --time=24:00:00 --mem=60g"