HUMAnN

HUMAnN is a pipeline for efficiently and accurately profiling the presence/absence and abundance of microbial pathways in a community from metagenomic or metatranscriptomic sequencing data. This process (functional profiling) aims to describe the metabolic potential of a microbial community and its members.

HUMAnN
- License
- Available
- Usage
  - Databases
  - Running HUMAnN
- Support
- More information

License

Free to use and open source under MIT License.

Available

Roihu-CPU: 4.0.0a2 (a HUMAnN 4.0 alpha release), via the bio-apps module.

Usage

HUMAnN is part of the bio-apps collection on Roihu. Load the bio-apps module tree and then the HUMAnN module:

module load bio-apps/v202603
module load humann/4.0.0a2

You can then run HUMAnN:

humann --help

Databases

HUMAnN's reference databases (ChocoPhlAn and UniRef) are several gigabytes and are not bundled with the module. HUMAnN also uses MetaPhlAn for taxonomic prescreening, which needs its own database.

Shared reference databases

CSC plans to provide shared reference databases at a central location on Roihu. This is still being set up. Until it is available, download and use your own copies as shown below.

HUMAnN stores its database paths in a configuration file (humann.cfg) inside the module's installation directory. On Roihu this installation is read-only, so humann_config and the default config update of humann_databases cannot save paths there. Instead, download the databases to a writable location (for example your project's /scratch directory) with --update-config no, and point HUMAnN at them on the command line for each run.

humann_databases --download chocophlan full /scratch/<project>/humann_db --update-config no
humann_databases --download uniref uniref90_diamond /scratch/<project>/humann_db --update-config no

Running HUMAnN

HUMAnN can utilize several CPU cores. Set --cpus-per-task to the desired number and pass --threads $SLURM_CPUS_PER_TASK to HUMAnN so it matches the requested number of cores. Point HUMAnN at your downloaded databases with --nucleotide-database and --protein-database (and MetaPhlAn's database via --metaphlan-options).

Example batch job script:

#!/bin/bash
#SBATCH --job-name=humann
#SBATCH --account=<project>
#SBATCH --output=output_%j.txt
#SBATCH --error=errors_%j.txt
#SBATCH --partition=small
#SBATCH --time=01:00:00
#SBATCH --nodes=1
#SBATCH --ntasks=1
#SBATCH --cpus-per-task=10
#SBATCH --mem-per-cpu=2000M

module load bio-apps/v202603
module load humann/4.0.0a2

DB=/scratch/<project>/humann_db

# Download a test file
wget https://github.com/biobakery/humann/raw/master/examples/demo.fastq.gz

# Run HUMAnN, specifying the database locations on the command line
humann --threads $SLURM_CPUS_PER_TASK --input demo.fastq.gz \
    --nucleotide-database $DB/chocophlan \
    --protein-database $DB/uniref \
    --output demo_out

Replace <project> with your CSC project (for example project_2001234), and use the same project in the database paths above.

Support

CSC Service Desk