VirSorter2

This article is a draft

This is not a complete article: This is a draft, a work in progress that is intended to be published into an article, which may or may not be ready for inclusion in the main wiki. It should not necessarily be considered factual or authoritative.

VirSorter2 is a tool to identify new viral sequences.

This page discusses how to install and use VirSorter2 v2.2.4.

Source code and documentation for VirSorter2 can be found on their GitHub page.

Remember to cite VirSorter2 if you use it for your analyses.

Installing VirSorter2 in a Python virtual environment

These instructions install VirSorter2 in your $HOME directory using Alliance's prebuilt Python wheels. Custom Python wheels are stored in /cvmfs/soft.computecanada.ca/custom/python/wheelhouse/. To install a VirSorter2 wheel, we will use the pip command and install it into a Python virtual environment.

1. Load the necessary modules.

[name@server ~]$ module load python/3.8 hmmer/3.3.2 prodigal/2.6.3

2. Create and activate a Python virtual environment.

[name@server ~]$ virtualenv --no-download ~/ENV_virsorter
[name@server ~]$ source ~/ENV_virsorter/bin/activate

3. Install VirSorter2 v2.2.4 in the virtual environment.

(ENV_virsorter) [name@server ~] pip install --no-index --upgrade pip
(ENV_virsorter) [name@server ~] pip install --no-index virsorter==2.2.4

4. Validate it.

(ENV_virsorter) [name@server ~] virsorter -h

5. Freeze the environment and requirements set.

(ENV_virsorter) [name@server ~] pip freeze > ~/virsorter-2.2.4-requirements.txt

6. Download the database in $SCRATCH with the --skip-deps-install option to bypass conda installation and also because dependencies are already installed.

(ENV_virsorter) [name@server ~] virsorter setup --db-dir $SCRATCH/db -j 4 --skip-deps-install

Testing VirSorter2

1. Download the test dataset in $SCRATCH.

(ENV_virsorter) [name@server ~] wget -O $SCRATCH/test.fa https://raw.githubusercontent.com/jiarong/VirSorter2/master/test/8seq.fa

2. Create a submission script

File : test-virsorter.sh

#!/bin/bash

#SBATCH --time=00:30:00
#SBATCH --mem-per-cpu=2G
#SBATCH --cpus-per-task=2

# Load modules dependencies
module load python/3.8 hmmer/3.3.2 prodigal/2.6.3

# Generate your virtual environment in $SLURM_TMPDIR
virtualenv --no-download $SLURM_TMPDIR/ENV
source $SLURM_TMPDIR/ENV/bin/activate
pip install --no-index --upgrade pip

# Install VirSorter2 and its dependencies
pip install --no-index -r ~/virsorter-2.2.4-requirements.txt

# Run VirSorter2 with the test dataset, using at most $SLURM_CPUS_PER_TASK and ignore conda.
# The database must already exist and you must specify its location.
virsorter run -w $SCRATCH/test.out -i $SCRATCH/test.fa --min-length 1500 -j $SLURM_CPUS_PER_TASK --verbose --use-conda-off --db-dir $SCRATCH/db all

3. If needed, start an interactive job to make sure the submission script launches correctly.

(ENV_virsorter) [name@server ~] {{{1}}}

salloc: Granted job allocation 1234567
$ bash test-virsorter.sh             # Run the submission script
$ exit            # terminate the allocation
salloc: Relinquishing job allocation 1234567

When the submission script works in an interactive job, you can complete the test run in a non-interactive job. 4. Then, submit the job to the scheduler.

(ENV_virsorter) [name@server ~] sbatch --job-name test-virsorter2 test-virsorter.sh

Upon a successful test run, you can run VirSorter2 using your own dataset.