MetaPhlAn

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.

MetaPhlAn is a "computational tool for profiling the composition of microbial communities (Bacteria, Archaea and Eukaryotes) from metagenomic shotgun sequencing data (i.e. not 16S) with species-level. With StrainPhlAn, it is possible to perform accurate strain-level microbial profiling", according to its GitHub repository. While the software stack on our clusters does contain modules for a couple of older versions (2.2.0 and 2.8) of this software, we now expect users to install recent versions using a Python virtual environment.

For more information on how to use MetaPhlan, see their wiki

Available wheels

You can list available wheels using the avail_wheels command:

[name@server ~]$ avail_wheels metaphlan --all-versions
name       version    python    arch
---------  ---------  --------  -------
MetaPhlAn  4.0.3      py3       generic
MetaPhlAn  3.0.7      py3       generic

Downloading databases

Note that MetaPhlAn requires a set of databases to be downloaded into the $SCRATCH.

Important: The database must live in the $SCRATCH

Databases can be downloaded from Segatalab FTP .

1. From a login node, create the data folder:

[name@server ~]$ export DB_DIR=$SCRATCH/metaphlan_databases
[name@server ~]$ mkdir -p $DB_DIR
[name@server ~]$ cd $DB_DIR

2. Download the data:

[name@server ~]$ parallel wget ::: http://cmprod1.cibio.unitn.it/biobakery4/metaphlan_databases/mpa_vJan21_CHOCOPhlAnSGB_202103.tar http://cmprod1.cibio.unitn.it/biobakery4/metaphlan_databases/mpa_vJan21_CHOCOPhlAnSGB_202103_marker_info.txt.bz2 http://cmprod1.cibio.unitn.it/biobakery4/metaphlan_databases/mpa_vJan21_CHOCOPhlAnSGB_202103_species.txt.bz2

Note that this step cannot be done from a compute node but must be done from a login node.

3. Extract the downloaded data, for example using an interactive job:

[name@server ~]$ salloc --account=<your account> --cpus-per-task=2 --mem=10G

Untar and unzip the databases:

[name@server ~]$ tar -xf mpa_vJan21_CHOCOPhlAnSGB_202103.tar
[name@server ~]$ parallel bunzip2 ::: *.bz2

Running MetaPhlAn

Once the database files have been downloaded and extracted, you can submit a job to initialize this database. You may edit the following job submission script according to your needs,

File : metaphlan-job.sh

#!/bin/bash

#SBATCH --account=def-someuser
#SBATCH --time=01:00:00
#SBATCH --cpus-per-task=4        # Number of cores
#SBATCH --mem=15G                # requires at least 15 GB of memory

# Load the required modules
module load gcc blast samtools bedtools bowtie2 python/3.10

# Move to the scratch
cd $SCRATCH

DB_DIR=$SCRATCH/metaphlan_databases

# Generate your virtual environment in $SLURM_TMPDIR
virtualenv --no-download ${SLURM_TMPDIR}/env
source ${SLURM_TMPDIR}/env/bin/activate

# Install metaphlan and its dependencies
pip install --no-index --upgrade pip
pip install --no-index metaphlan==X.Y.Z  # EDIT: the required version here, e.g. 4.0.3

# Reuse the number of core allocated to our job from `--cpus-per-task=4`
# It is important to use --index and --bowtie2db so that MetaPhlAn can run inside the job
metaphlan metagenome.fastq --input_type fastq -o profiled_metagenome.txt -nproc $SLURM_CPUS_PER_TASK --index mpa_vJan21_CHOCOPhlAnSGB_202103 --bowtie2db $DB_DIR --bowtie2out metagenome.bowtie2.bz2