BiodivPipeline

A modular nf-core / Nextflow pipeline that turns raw biodiversity CSV records into a quality-annotated, taxonomically resolved dataset and FAIR RDF triples.

Framework: nf-core template v4.0.2

Quick start

Prerequisites: Java 11+, Nextflow >= 25.10.4, Docker running.

# 1. Set the module API keys (what each is for: docs/runbook/running.md#secrets)
nextflow secrets set WP2_BIODIV_API_KEY '<BiodivPortal key>'
nextflow secrets set WP3_OPENAI_API_KEY '<OpenAI key>'
nextflow secrets set WP4_OPENAI_API_KEY '<OpenAI key>'
nextflow secrets set WP5_OPENAI_API_KEY '<OpenAI key>'

# 2. Run the bundled test dataset
nextflow run main.nf -profile test,docker --outdir results

# 3. Run your own data
nextflow run main.nf -profile docker --input path/to/biodiv.csv --outdir results

No WP5 key? Add --skip_provisional_concepts.

Full run guide, secrets, and parameters: docs/runbook/running.md.

Project structure

main.nf                       # entry point
workflows/biodivpipeline.nf   # workflow DAG
modules/local/                # one module per work package
conf/                         # config profiles
docs/                         # documentation (see below)

Documentation

Start at the documentation index:

• Runbook — run, operate, troubleshoot
• Development — contribute and add modules
• Reference — architecture and module interfaces

Built on nf-core — see CITATIONS.md.