slambrechts / virsearch

Snakemake workflow to detect and classify viruses in metagenome assemblies.

VirSearch - searching viral sequences in metagenomes

Snakemake workflow to detect and classify viruses in metagenome assemblies.

It first detects viral sequences in assemblies (.fa files) with VirSorter2, VIBRANT and DeepVirFinder. Predictions are strictly quality controlled with CheckV, followed by clustering with CD-HIT and taxonomic classification with Demovir.

Installation

1. Install conda, snakemake and USEARCH.

2. Clone repository

git clone --recursive https://github.com/alexmsalmeida/virsearch.git

3. Download and extract necessary databases (uncompressed directory will require a total of 30 GB).

wget http://ftp.ebi.ac.uk/pub/databases/metagenomics/genome_sets/virsearch_db.tar.gz
tar -xzvf virsearch_db.tar.gz

How to run

1. Edit config.yml file to point to the input, output and databases directories, as well as the USEARCH binary location (usearch_binary). Input directory should contain the .fa assemblies to analyse.

2. (option 1) Run the pipeline locally (adjust -j based on the number of available cores)

snakemake --use-conda -k -j 4

2. (option 2) Run the pipeline on a cluster (e.g., LSF)

snakemake --use-conda -k -j 100 --cluster-config cluster.yml --cluster 'bsub -n {cluster.nCPU} -M {cluster.mem} -o {cluster.output}'

Output

The main output files generated per input FASTA are the final_predictions.fa and final_predictions_tax.tsv files, which contain the viral sequences in FASTA format and their taxonomic annotation, respectively. If these files are empty it likely means that no high-confidence viral sequences were detected (check individual logs of the tools to confirm no other issues arose).