OrthoExplorer

OrthoExplorer is a modular Nextflow DSL2 pipeline for preparing proteomes, inferring orthogroups with OrthoFinder, and performing downstream gene-conservation and pangenome analyses.

Workflow

The pipeline is divided into three subworkflows:

1. Dataset preparation

   • validates species and outgroup tables;
   • uses user-provided proteomes when available;
   • otherwise downloads a RefSeq genome and GFF3 from NCBI;
   • extracts the longest protein isoform for each gene with AGAT;
   • downloads the NCBI taxonomy for all taxa.

2. Orthology inference

   • collects all prepared proteomes;
   • runs OrthoFinder;
   • exports normalized OrthoFinder result filenames.

3. Downstream analyses

   • reformats OrthoFinder species names;
   • assigns gene-conservation ranks;
   • defines pangenome categories;
   • generates Venn/UpSet summaries.

Input tables

Both species.csv and outgroups.csv must contain the following columns:

taxid,name,file
126957,Strigamia maritima,/path/to/strigamia-maritima.faa
1490507,Geophilus flavus,

taxid and name are mandatory and must be non-empty.

file is optional:

• when populated, the referenced protein FASTA is used;
• when empty, OrthoExplorer downloads a RefSeq genome and GFF3 and extracts one canonical protein isoform per gene.

Paths in file should be absolute, or valid relative to the directory from which Nextflow is launched.

Required software

The current modules expect these commands to be available through the execution environment:

• Nextflow;
• NCBI Datasets CLI;
• unzip;
• AGAT;
• OrthoFinder;
• download_taxonomy.py;
• reformat_orthofinder_tables.py (included in bin/);
• assign_gene_conservation_rank.py;
• venn_upset_orthogroups.py.

Software environments should preferably be defined with containers or Conda in the Nextflow configuration rather than with hard-coded module load commands inside process modules.

Configuration

Example configuration:

params {
    species   = "${projectDir}/species_myriapoda.csv"
    outgroups = "${projectDir}/outgroups_myriapoda.csv"
    groups    = "Diplopoda,Chilopoda,Hexapoda,Crustacea,Chelicerata"
    colors    = "${projectDir}/data/colors.yaml"

    outdir    = "${projectDir}/results"
    cache_dir = "${projectDir}/cache/work"
    conda_env = "${projectDir}/envs/orthoexplorer"
}

workDir = params.cache_dir

Process resources are defined in conf/modules.config, while publication rules are defined in conf/publish.config.

Running the pipeline

nextflow run main.nf \
    -c nextflow_myriapoda.config \
    -resume

Parameters may also be overridden from the command line:

nextflow run main.nf \
    -c nextflow_myriapoda.config \
    --species species_myriapoda.csv \
    --outgroups outgroups_myriapoda.csv \
    --colors data/colors.yaml \
    --outdir results \
    -resume

Stub testing

Each process contains a stub block that checks whether its main command is available and creates the expected empty outputs.

nextflow run main.nf \
    -c nextflow_myriapoda.config \
    -stub-run

The stub run validates workflow wiring and output declarations, but it does not validate the biological content of the outputs.

Output structure

results/
├── taxonomy/
├── orthofinder/
├── gene_conservation/
└── pangenome/

The Nextflow execution cache is stored separately under cache/work/.

---

Credits

Originally written by Dorine Merlat (dorine.merlat@etu.unistra.fr). Thanks to Arnaud Kress and Odile Lecompte for assistance.

Citations

An extensive list of references for the tools used by the pipeline can be found in the CITATIONS.md file.

---

License

This project is distributed under the MIT License.