PhyloTwin-Preprocessor - GBIF Occurrence Data Processing Pipeline

A Nextflow pipeline for processing GBIF species occurrence data with spatial outlier detection and H3 grid binning.

The pipeline offers two processing modes:

• "Atomic" mode: Processes each species independently
• "Batched" mode: Processes multiple species in batches (optimized for HPC environments)

Dependencies

Primary dependencies

These are the tools that the user must install and configure on their system to run the pipeline:

• Nextflow >= 24.10 (requires Java >= 17 & <= 23)
• Singularity/Apptainer or Docker

Secondary dependencies

These are the containerized tools or packages required by the pipeline, which will be automatically handled within the containers:

• DuckDB
• ELKI
• R
• Python
• GNU Parallel
• aria2
• jq

R packages:

• data.table
• ape