Parasail

parasail is a SIMD C (C99) library containing implementations of the Smith-Waterman (local), Needleman-Wunsch (global), and various semi-global pairwise sequence alignment algorithms.

Starting from StdEnv/2023, parasail python extension is now bundled in the module parasail. As for 2020, the module needs to be loaded in order for the python extension to be installed in a virtual environment.

Usage

One can find the available version using

[name@server ~]$ module spider parasail

and load the library using

[name@server ~]$ module load parasail/2.6.2

parasail_aligner Example

When using the binary parasail_aligner it is important to set the number of threads according to the number of cores allocated in our job. We can set it with

parasail_aligner -t ${SLURM_CPUS_PER_TASK:-1} ...}}

Python extension

The module contains bindings for multiple Python versions. To discover which are the compatible Python versions, run

[name@server ~]$ module spider parasail/1.3.4

Usage

1. Load the required modules.

[name@server ~]$ module load parasail/2.6.2 python/3.11 scipy-stack/2023b

2. Import parasail 1.3.4.

[name@server ~]$ python -c "import parasail"

If the command displays nothing, the import was successful.

Example

Run a quick local alignment score comparison between BioPython and parasail.

1. Write the python script:

File : parasail-sw.py

import parasail
from Bio.Align import PairwiseAligner

A = "ACGT" * 1000

# parasail
matrix = parasail.matrix_create("ACGT", 1, 0)
parasail_score = parasail.sw(A, A, 1, 1, matrix).score

# biopython
bio_score = PairwiseAligner().align(A, A)[0].score

print('parasail:', parasail_score)
print('biopython:', bio_score)

2. And the job submission script:

Default StdEnvStdEnv/2020

File : submit-parasail.sh

#!/bin/bash
#SBATCH --account=def-someuser  # replace with your PI account
#SBATCH --cpus-per-task=1 
#SBATCH --mem-per-cpu=3G      # increase as needed
#SBATCH --time=1:00:00

module load parasail/2.6.2 python/3.11 scipy-stack/2023b

# Install any other requirements, such as Biopython
virtualenv --no-download $SLURM_TMPDIR/env
source $SLURM_TMPDIR/env/bin/activate
pip install --no-index --upgrade pip
pip install --no-index biopython==1.83

python parasail-sw.py

2.1. Identify available wheels first :

[name@server ~]$ avail_wheel parasail

Then install the desired version in your virtual environment:

File : submit-parasail.sh

#!/bin/bash
#SBATCH --account=def-someuser  # replace with your PI account
#SBATCH --cpus-per-task=1 
#SBATCH --mem-per-cpu=3G      # increase as needed
#SBATCH --time=1:00:00

module load StdEnv/2020 gcc parasail/2.5 python/3.10

# Install any other requirements, such as Biopython
virtualenv --no-download $SLURM_TMPDIR/env
source $SLURM_TMPDIR/env/bin/activate
pip install --no-index --upgrade pip
pip install --no-index parasail==1.2.4 biopython==1.83

python parasail-sw.py

3. Then submit the job with

[name@server ~]$ sbatch submit-parasail.sh
name      version    python    arch
--------  ---------  --------  -------
parasail  1.2.4      py2,py3   generic

4. The output will be in the slurm output file, once the job has run:

[name@server ~]$ less slurm-*.out
parasail: 4000
biopython: 4000.0

Available Python packages

Other Python packages that depend on parasail will have their requirement satisfied with the module loaded:

[name@server ~]$ pip list | grep parasail
parasail                           1.3.4