UoS eDNA Bioinformatics Pipeline

An automated environmental DNA (eDNA) metabarcoding pipeline that converts raw NCBI SRA sequencing data into taxonomically annotated species tables.

This pipeline was created as a collaboration between the Genome Damage and Stability Centre (GDSC), and Ecology and Evolution (EE) department at the University of Sussex (UoS). It is built for research and meant to serve as an alternative to external FASTQ analysis.

Currently only works for 12S MiFish-U and 16S MarVer3 Primers.

The upload here is hardcoded to the example dataset "SR normalised Clark et al 2024" (Clark, et al,. 2024) but the file location is easily adapatable.

• If FASTQ files already exist, create a folder named fqconvert in your working directory and place them in there and ensure they end in _1.fastq or _2.fastq.

• If working with similar archived files change line 16 TARGET_NAME="SR normalised Clark et al 2024" of 00_sra_to_fastq.sh to your chosen folder and ensure it is in the same directory as the pipeline.

After running your outputs can be traced and validated at every stage through the folders in the generated runs folder. A run will be created with the date of running the pipeline.

If only interested in the final report, type master_species_composition_taxonomy.csv in your root directory.

---

Repository contents

This repository includes:

• The complete eDNA metabarcoding pipeline for MiFish-U 12S and MarVer3 16S primers
• Benchmark NCBI SRA sequencing data
• An example of the output species table (master_species_composition.csv)
• The original Nature Metrics reference dataset (reference_data.xlsx)

---

Requirements

This pipeline was validated on WSL Ubuntu and built on Ubuntu and the Linux Rocky 9 Artemis High-Performance-Computing (HPC) Cluster. Your experience on other operating systems may vary.

Install:

• Python 3.11+
• SRA Toolkit
• Cutadapt
• USEARCH v12+

Then install the Python packages:

pip install -r requirements.txt

---

Running the pipeline

Make the launcher executable:

chmod +x run_pipeline.sh

Run the complete workflow:

./run_pipeline.sh

The pipeline automatically performs:

1. SRA → FASTQ conversion
2. Primer trimming
3. Dereplication
4. Denoising
5. BLAST identification
6. NCBI taxonomy annotation

---

Output

A new results folder is created inside:

runs/

The final output file is:

master_species_composition.csv

On supported Linux systems, this file opens automatically when the pipeline finishes.

---

Included example files

If you simply wish to inspect the benchmark results without running the pipeline, the repository already includes:

• master_species_composition.csv — the completed taxonomic annotation generated by the pipeline.
• reference_data.xlsx — the original Nature Metrics reference report used for comparison and validation.

---

Citation

If you use this pipeline in research or publications, please cite:

Rudie Kauhanen. UoS eDNA Bioinformatics Pipeline version 1.0., 2026. https://github.com/Rudie-K/UoS-eDNA-Bioinformatics

Please also cite the software used by this workflow where appropriate, including:

• Cutadapt
• USEARCH
• NCBI BLAST
• NCBI Taxonomy