SuppOrth (Support-based Orthologue integration) is a lightweight, flexible Python tool to unify and compare orthologue predictions across multiple tools.
It aggregates orthologue calls from various predictors, enriches them with BLAST alignment data, and optionally collapses isoform-level matches to gene-level orthology relationships.
- Combine results from multiple orthologue predictors (e.g., Broccoli, SonicParanoid, OrthoFinder,OMA etc.)
- Support-aware integration of orthologue calls
- Use BLAST identity and e-value for additional scoring
- Optional collapsing of isoforms using protein-to-gene mapping
- JSON and Pickle support for input formats
The script is not constrained to any specific tool, but it is recomended to combine variety of aligner, clustering stratgies to improve the consensus support.
Here we use:
Run the tool with specific predictor files:
python suppOrth.py -p predictors.json,predictors2.json -l B,F -B blast1.pickle,blast2.pickle -o output.tsv
Use bulk mode to automatically load all predictor JSONs in a folder:
python suppOrth.py --bulk predictors_folder/ -B blast1.pickle,blast2.pickle -o output.tsv
To collapse to gene-level (if isoforms exist and mapping dictionaries are provided):
python suppOrth.py ... -p2g prot2gene_species1.pickle,prot2gene_species2.pickle
Tab-separated file (.tsv) including:
- Protein or gene pairs
- Level of support
- Supporting tools
- BLAST identity & e-value
- (Optional) Aggregated gene-level support
