Tutorial: Miniforge setup for bioinformatics

This is a short, practical guide for new lab members setting up Miniforge on Linux or macOS.

Official references:

• Miniforge docs and releases: https://github.com/conda-forge/miniforge

1) Check your architecture

uname
uname -m

Expected architecture values:

• Linux Intel/AMD: x86_64
• Linux ARM64: aarch64
• macOS Intel: x86_64
• macOS Apple Silicon (M1/M2/M3/M4 and newer): arm64

2) Download and run the official installer

Linux (recommended command from Miniforge docs)

wget -O Miniforge3.sh "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-$(uname)-$(uname -m).sh"
bash Miniforge3.sh

macOS (recommended command from Miniforge docs)

curl -fsSLo Miniforge3.sh "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-MacOSX-$(uname -m).sh"
bash Miniforge3.sh

During interactive install:

• use default install location (recommended)
• allow installer to run conda init

Then close and reopen terminal.

3) Confirm installation

conda --version
mamba --version

If commands are not found:

~/miniforge3/bin/conda init

Restart terminal after that.

4) Create your lab environment

mamba create -n lab python=3.11 -y
conda activate lab
mamba install -y numpy pandas scipy matplotlib jupyterlab

5) Daily usage

conda activate lab
conda env list
conda deactivate

6) Best practice for lab work

• Do not install project dependencies into base.
• Use one conda environment per project.
• Export an environment file when sharing workflows:

conda env export --no-builds > environment.yml

Recreate the same environment elsewhere:

mamba env create -f environment.yml

7) Optional cleanup / shell behavior

Disable auto-activating base at every terminal start:

conda config --set auto_activate_base false

If shell activation is still broken:

conda init bash   # Linux bash users
conda init zsh    # macOS default shell

Then restart terminal.