Managing Conda Environments on the HPC

SOP ID: HPC-002 Version: 1.0 Date: 2026-02-23 Author: AGRP

Overview

Conda is a package and environment manager that lets you install bioinformatics software without needing administrator access. Each environment is an isolated space with its own set of tools and versions — so different projects never conflict with each other.

Full Course Available

This SOP covers the essentials. For a comprehensive walkthrough including advanced workflows and best practices, see the full course:

Conda for Bioinformatics

1. Installation

Each user installs conda in their own home directory. Download and run the Miniconda installer:

wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
bash Miniconda3-latest-Linux-x86_64.sh

Accept the defaults and allow the installer to initialise conda in your .bashrc. Then reload your shell:

source ~/.bashrc

Confirm it works:

conda --version

2. First-Time Setup

Add the bioinformatics channels so that tools like GATK, BWA, and QIIME2 are available:

conda config --add channels defaults
conda config --add channels bioconda
conda config --add channels conda-forge

3. Creating an Environment

Always create a dedicated environment for each project or tool rather than installing everything into base.

conda create -n my_project python=3.11

Activate it:

conda activate my_project

Your prompt will update to show the active environment:

(my_project) your_username@lab417:~$

4. Installing Software

With your environment active, install tools using conda install:

conda install -c bioconda samtools bwa

Search for a package:

conda search -c bioconda fastqc

5. Essential Commands

Command	What it does
`conda activate my_env`	Switch into an environment
`conda deactivate`	Return to base
`conda env list`	List all your environments
`conda list`	List packages in the active environment
`conda install package`	Install a package
`conda remove package`	Remove a package
`conda env remove -n my_env`	Delete an environment entirely

Export your environment so collaborators (or your future self) can recreate it exactly:

conda env export > environment.yml

Recreate it from the file:

conda env create -f environment.yml

7. Using Conda in SLURM Jobs

Activate your environment inside your SLURM batch script:

#!/bin/bash
#SBATCH --job-name=my_analysis
#SBATCH --partition=agrp
#SBATCH --cpus-per-task=4
#SBATCH --mem=8G

source ~/miniconda3/etc/profile.d/conda.sh
conda activate my_project

# Your commands here
python my_script.py

Warning

Do not rely on your .bashrc to activate conda inside batch jobs — always source it explicitly as shown above.