pharmkg

Name: pharmkg
Author: QSong-github

132stars4forks

How to install pharmkg?

qsong-github/drugclaw/pharmkg

$npx -y skills add QSong-github/DrugClaw --skill pharmkg

Installs into the current project.

Summary

Query the PharmKG knowledge graph (180k entities, 39 relation types, >1M triples). Use whenever the user asks about biomedical relationships among genes, drugs/chemicals, and diseases — e.g. drug–gene interactions, drug–disease associations, gene–disease links, or drug–drug relat

Files · 1

View on GitHub

SKILL.md

1	# PharmKG Query Skill
2
3	Search PharmKG triples by entity name. Matching is case-insensitive; exact match
4	is tried first via a prebuilt index, with substring fallback.
5
6	\| Input Example \| Matches On \|
7	\|---\|---\|
8	\| `aspirin` \| exact on Entity1\_name or Entity2\_name \|
9	\| `BRCA1` \| exact / substring on entity names \|
10	\| `Alzheimer Disease` \| substring on entity names \|
11
12	## Entity & Relation Types
13
14	Entities (~188 k): Drug/Chemical (DrugBank, ChEMBL), Gene/Protein (Entrez, UniProt), Disease (DO, MeSH).
15
16	Relations (39 types): chemical–gene (inhibition, activation, binding …), chemical–disease (treatment, marker, risk factor), gene–disease (association, marker), chemical–chemical (similarity, interaction), and others.
17
18	## API
19
20	\| Function \| Input \| Returns \|
21	\|---\|---\|---\|
22	\| `load_pharmkg(path)` \| CSV path \| `list[dict]` of triples \|
23	\| `_build_index(triples)` \| triple list \| `dict[str, list]` (entity→triples) \|
24	\| `search(triples, entity, index=)` \| entity string \| `list[dict]` \|
25	\| `search_batch(triples, entities, index=)` \| list of strings \| `dict[str, list]` \|
26	\| `summarize(hits, entity)` \| hits + label \| compact text \|
27	\| `to_json(hits)` \| hits \| JSON string \|
28
29	## Usage
30
31	See `if __name__ == "__main__"` block in `55_PharmKG.py` for runnable examples
32	covering: single entity search, batch search, summarize, and JSON output.
33
34	## Data
35
36	- Source: `raw_PharmKG-180k.csv` (comma-separated)
37	- Columns: `Entity1_name`, `relationship_type`, `Entity2_name`
38	- Scale: ~188 k entities, 39 relation types, >1 M triples
39	- Path: `DATA_PATH` variable in `55_PharmKG.py`
40	- Paper: Zheng et al., Briefings in Bioinformatics 22(4), 2021. DOI: 10.1093/bib/bbaa344

People who liked this also viewed

6 picks

Type

pharmkg

132stars4forks

View on GitHub

How to install pharmkg?

qsong-github/drugclaw/pharmkg

$npx -y skills add QSong-github/DrugClaw --skill pharmkg

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# PharmKG Query Skill
2
3	Search PharmKG triples by entity name. Matching is case-insensitive; exact match
4	is tried first via a prebuilt index, with substring fallback.
5
6	\| Input Example \| Matches On \|
7	\|---\|---\|
8	\| `aspirin` \| exact on Entity1\_name or Entity2\_name \|
9	\| `BRCA1` \| exact / substring on entity names \|
10	\| `Alzheimer Disease` \| substring on entity names \|
11
12	## Entity & Relation Types
13
14	Entities (~188 k): Drug/Chemical (DrugBank, ChEMBL), Gene/Protein (Entrez, UniProt), Disease (DO, MeSH).
15
16	Relations (39 types): chemical–gene (inhibition, activation, binding …), chemical–disease (treatment, marker, risk factor), gene–disease (association, marker), chemical–chemical (similarity, interaction), and others.
17
18	## API
19
20	\| Function \| Input \| Returns \|
21	\|---\|---\|---\|
22	\| `load_pharmkg(path)` \| CSV path \| `list[dict]` of triples \|
23	\| `_build_index(triples)` \| triple list \| `dict[str, list]` (entity→triples) \|
24	\| `search(triples, entity, index=)` \| entity string \| `list[dict]` \|
25	\| `search_batch(triples, entities, index=)` \| list of strings \| `dict[str, list]` \|
26	\| `summarize(hits, entity)` \| hits + label \| compact text \|
27	\| `to_json(hits)` \| hits \| JSON string \|
28
29	## Usage
30
31	See `if __name__ == "__main__"` block in `55_PharmKG.py` for runnable examples
32	covering: single entity search, batch search, summarize, and JSON output.
33
34	## Data
35
36	- Source: `raw_PharmKG-180k.csv` (comma-separated)
37	- Columns: `Entity1_name`, `relationship_type`, `Entity2_name`
38	- Scale: ~188 k entities, 39 relation types, >1 M triples
39	- Path: `DATA_PATH` variable in `55_PharmKG.py`
40	- Paper: Zheng et al., Briefings in Bioinformatics 22(4), 2021. DOI: 10.1093/bib/bbaa344

People who liked this also viewed

6 picks

Type