Data Setup

Before working through the workshop exercises, you need to download the sequencing data, reference genome, and build the STAR genome index. This page walks through those steps with a single setup script.

Overview

We will use six paired-end RNA-seq samples from a 2025 Saccharomyces cerevisiae study investigating whether phosphorylation of the Arp8 subunit of the INO80 chromatin remodeler complex by Dbf4-Dependent Kinase (DDK) influences genome-wide transcription. Gene expression profiles from isogenic Arp8 wild-type and arp8-P (phospho-mutant: S65A/S233A) strains are compared across three biological replicates per condition. The dataset is available from BioProject PRJNA1185750 and published in Nature Communications (2026).

The yeast genome is approximately 12 MB, small enough for the STAR index to build in minutes and for alignment to finish in seconds per sample, making it ideal for a classroom setting.

Sample name	Condition	SRA accession
`wt_rep1`	Arp8 wild-type replicate 1	SRR31334598
`wt_rep2`	Arp8 wild-type replicate 2	SRR31334596
`wt_rep3`	Arp8 wild-type replicate 3	SRR31334594
`mut_rep1`	arp8-P mutant replicate 1	SRR31334597
`mut_rep2`	arp8-P mutant replicate 2	SRR31334595
`mut_rep3`	arp8-P mutant replicate 3	SRR31334593

Expected Directory Structure

By the end of this page, your project directory should look like this:

snakemake_tutorial/
├── data/
│   ├── wt_rep1_R1.fastq.gz
│   ├── wt_rep1_R2.fastq.gz
│   ├── wt_rep2_R1.fastq.gz
│   ├── wt_rep2_R2.fastq.gz
│   ├── wt_rep3_R1.fastq.gz
│   ├── wt_rep3_R2.fastq.gz
│   ├── mut_rep1_R1.fastq.gz
│   ├── mut_rep1_R2.fastq.gz
│   ├── mut_rep2_R1.fastq.gz
│   ├── mut_rep2_R2.fastq.gz
│   ├── mut_rep3_R1.fastq.gz
│   └── mut_rep3_R2.fastq.gz
├── genome/
│   ├── Saccharomyces_cerevisiae.R64-1-1.dna.fa
│   ├── Saccharomyces_cerevisiae.R64-1-1.gtf
│   └── star_index/
│       └── (STAR index files)
└── Snakefile

Step 1: Install Setup Tools

Activate your snakemake conda environment and install the tools needed to download and prepare the data:

conda activate snakemake_workshop
mamba install -c bioconda -c conda-forge sra-tools

Step 2: Run the Setup Script

Save the following as setup.sh in your project directory, then run it. The script downloads the full FASTQ files from SRA, downloads the S. cerevisiae R64 reference genome and GTF from Ensembl, and builds the STAR genome index.

#!/usr/bin/env bash
set -euo pipefail

echo "==> Creating project directories..."
mkdir -p data genome/star_index results

# ─── Download FASTQ files from SRA ───────────────────────────────────────────
declare -A SAMPLES=(
    ["wt_rep1"]="SRR31334598"
    ["wt_rep2"]="SRR31334596"
    ["wt_rep3"]="SRR31334594"
    ["mut_rep1"]="SRR31334597"
    ["mut_rep2"]="SRR31334595"
    ["mut_rep3"]="SRR31334593"
)

echo "==> Downloading FASTQ files from SRA (this may take several minutes per sample)..."
for NAME in "${!SAMPLES[@]}"; do
    ACC="${SAMPLES[$NAME]}"
    echo "    Fetching ${ACC} -> ${NAME}..."
    fasterq-dump "${ACC}" \
        --outdir data/ \
        --split-files \
        --threads 4 \
        --progress

    # Compress and rename to workshop convention
    gzip -c "data/${ACC}_1.fastq" > "data/${NAME}_R1.fastq.gz"
    gzip -c "data/${ACC}_2.fastq" > "data/${NAME}_R2.fastq.gz"

    # Remove uncompressed SRA downloads
    rm -f "data/${ACC}_1.fastq" "data/${ACC}_2.fastq"
    echo "    Done: ${NAME}"
done

# ─── Download Reference Genome and Annotation ────────────────────────────────
ENSEMBL="https://ftp.ensembl.org/pub/release-110"

echo "==> Downloading S. cerevisiae reference genome..."
wget -q --show-progress \
    -O genome/Saccharomyces_cerevisiae.R64-1-1.dna.fa.gz \
    "${ENSEMBL}/fasta/saccharomyces_cerevisiae/dna/Saccharomyces_cerevisiae.R64-1-1.dna.toplevel.fa.gz"
gunzip genome/Saccharomyces_cerevisiae.R64-1-1.dna.fa.gz

echo "==> Downloading GTF annotation..."
wget -q --show-progress \
    -O genome/Saccharomyces_cerevisiae.R64-1-1.gtf.gz \
    "${ENSEMBL}/gtf/saccharomyces_cerevisiae/Saccharomyces_cerevisiae.R64-1-1.110.gtf.gz"
gunzip genome/Saccharomyces_cerevisiae.R64-1-1.gtf.gz

# ─── Build STAR Genome Index ─────────────────────────────────────────────────
echo "==> Building STAR genome index (~3 minutes)..."
STAR \
    --runMode genomeGenerate \
    --runThreadN 4 \
    --genomeDir genome/star_index/ \
    --genomeFastaFiles genome/Saccharomyces_cerevisiae.R64-1-1.dna.fa \
    --sjdbGTFfile genome/Saccharomyces_cerevisiae.R64-1-1.gtf \
    --genomeSAindexNbases 10

echo ""
echo "==> Setup complete! Verifying files..."
echo "--- data/ ---"
ls -lh data/
echo "--- genome/ ---"
ls -lh genome/
echo ""
echo "You are ready to begin the workshop!"

Run it:

bash setup.sh

Expected runtime and disk usage

Step	Time	Space
SRA download (6 samples)	15–20 min	~4 GB temporary
FASTQ files (12 files)	—	~1.5 GB (100–144 MB each)
Reference genome + GTF	<1 min	~22 MB
STAR genome index	<1 min	~150 MB

Total setup time: ~15 minutes | Total space after cleanup: ~1.7 GB

Why --genomeSAindexNbases 10?

STAR's default suffix array parameter (14) is tuned for large genomes like human (~3 GB). For the S. cerevisiae genome (~12 MB), a value of 10 is required, otherwise STAR prints a warning and produces a suboptimal index.

Step 3: Verify Your Setup

Run the following to confirm all files are in place before starting the workshop:

# Should list 12 .fastq.gz files
ls data/

# Should list the .fa file, .gtf file, and star_index/
ls genome/

# Should contain SA, Genome, and other STAR index files
ls genome/star_index/

Once everything looks correct, open your project directory in your terminal and proceed to Snakemake Fundamentals.