Skip to content

Module 7 — Local Env + Tiny QC

Time: 60–75 min
Goal: Create a clean Conda/Mamba env and run seqtk, fastqc, and multiqc locally.

Type it, don't paste it

Muscle memory now = speed later. Please type once before copy/paste.

1) Install Miniforge/Mambaforge

Bioinformatics often requires many specialized tools, and managing their dependencies can be tricky. Conda (and its faster alternative, Mamba) helps create isolated environments for these tools, preventing conflicts and making your analyses reproducible. Mambaforge is a distribution that includes Mamba and Conda.

cd ~
wget https://github.com/conda-forge/miniforge/releases/latest/download/Mambaforge-Linux-x86_64.sh
bash Mambaforge-Linux-x86_64.sh -b -p $HOME/mambaforge
# This line sets up your shell to use conda/mamba commands
eval "$($HOME/mambaforge/bin/conda shell.bash hook)"
conda config --set auto_activate_base false

cd ~
# Use -L to follow redirects and a specific version for stability
curl -L -O https://github.com/conda-forge/miniforge/releases/download/24.7.1-0/Mambaforge-MacOSX-arm64.sh
# Verify the downloaded file is a shell script before running
file Mambaforge-MacOSX-arm64.sh
bash Mambaforge-MacOSX-arm64.sh -b -p $HOME/mambaforge
# This line sets up your shell to use conda/mamba commands
eval "$($HOME/mambaforge/bin/conda shell.zsh hook)"  # or bash if you use bash
conda config --set auto_activate_base false

cd ~
# Use -L to follow redirects and a specific version for stability
curl -L -O https://github.com/conda-forge/miniforge/releases/download/24.7.1-0/Mambaforge-MacOSX-x86_64.sh
# Verify the downloaded file is a shell script before running
file Mambaforge-MacOSX-x86_64.sh
bash Mambaforge-MacOSX-x86_64.sh -b -p $HOME/mambaforge
# This line sets up your shell to use conda/mamba commands
eval "$($HOME/mambaforge/bin/conda shell.zsh hook)"
conda config --set auto_activate_base false

2) Make the env

conda create -y -n rnaseq101 -c conda-forge -c bioconda fastqc multiqc seqtk
conda activate rnaseq101

3) Prepare a tiny FASTQ and downsample

mkdir -p ~/de-onramp/lesson4 && cd ~/de-onramp/lesson4
cat > toy.fq << 'EOF'
@r1
ACGTTGCA
+
IIIIHHHF
@r2
GGGTTTAA
+
FFFFFIII
@r3
ACGTACGT
+
IIIIIIII
EOF

# Downsample "reads" (works better on real data but fine for demo)
seqtk sample -s 1 toy.fq 1.0 > toy.sub.fq

4) Run FastQC + MultiQC

fastqc -o qc toy.fq toy.sub.fq
multiqc -o qc qc

Open qc/multiqc_report.html locally (VS Code file explorer → click to open in your default browser).

Exit Ticket (email)

Subject: DE M7 Exit Ticket –
Paste:

  • Output of conda list | grep -E 'fastqc|multiqc|seqtk'
  • Two bullets from your MultiQC report (e.g., per-base quality and GC content).