Oregano

Skill BioTender-max/awesome-bio-agent-skills/skills/drugclaw/oregano

A curated collection of AI agent skills for biomedical research, covering genomics, proteomics, single-cell analysis, clinical AI, and protein design.

Install

npx -y skills add BioTender-max/awesome-bio-agent-skills --skill oregano

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.

What its author says it does

Copied from the file, not written here

Query the OREGANO knowledge graph for computational drug repurposing. Use whenever the user asks about drug–target–disease–gene–pathway relationships, compound cross-references, drug repurposing hypotheses, or wants to explore neighbors of any biomedical entity in a knowledge graph that includes natural compounds.

SKILL.md

2.6 KB, as published. Nobody here has run it

OREGANO Query Skill

Search the OREGANO knowledge graph (88,937 nodes, 824,231 links) by any entity. Auto-resolves input to OREGANO node IDs via cross-reference tables.

Input Pattern	Detected As	Match Logic
OREGANO internal ID (e.g. `1234`)	OREGANO node ID	exact in triplet index
`DB00331` / DrugBank ID	external xref	exact in COMPOUND.tsv
UniProt / KEGG / MeSH / UMLS ID	external xref	exact across all metadata TSVs
`metformin`, `BRCA1`, free text	entity name	substring on name columns

API

Function	Input	Returns
`search(query)`	single entity string	dict: `{query, resolved_ids, metadata, triplets}`
`search_batch(queries)`	list of entity strings	`dict[str, search_result]`
`summarize(result)`	search result dict	compact LLM-readable text
`to_json(result)`	search result dict	JSON-serializable dict
`get_stats()`	—	graph-level counts (triplets, nodes, predicates, entity types)

Graph Schema

11 node types: Compound (90,868), Gene (35,794), Target (22,096), Disease (18,333), Phenotype (11,605), Side Effect (6,060), Indication (2,714), Pathway (2,129), Effect (171), Activity (78).

19 relation types (predicates): e.g. has_target, has_indication, has_side_effect, interacts_with, involved_in_pathway, associated_with, has_phenotype, etc. Run get_stats() to list all predicates with counts.

Usage

See if __name__ == "__main__" block in 21_OREGANO.py for runnable examples covering: free-text drug name search, DrugBank ID lookup, batch search, JSON pipeline output, and graph statistics.

Data

Source: Zenodo DOI 10.5281/zenodo.10103842 (CC-BY 4.0)
Version: v2.1 (published 2023-11-10)
Core file: OREGANO_V2.1.tsv — tab-delimited triplets (Subject, Predicate, Object)
Metadata files: COMPOUND.tsv, TARGET.tsv, GENES.tsv, DISEASES.tsv, PHENOTYPES.tsv, PATHWAYS.tsv, INDICATION.tsv, SIDE_EFFECT.tsv, ACTIVITY.tsv, EFFECT.tsv
Path: DATA_DIR variable in 21_OREGANO.py

Citation

Boudin, M., Diallo, G., Drancé, M. & Mougin, F. The OREGANO knowledge graph for computational drug repurposing. Sci Data 10, 871 (2023). https://doi.org/10.1038/s41597-023-02757-0