tooluniverse-kegg-disease-drug — Claude Code skill
tooluniverse-kegg-disease-drug is a Claude Code skill that links diseases to genes, drugs to targets and variants to pathways via KEGG. It activates on keywords like KEGG disease, drug targets or variant networks.
Apache-2.0 235 installs ★ 1,798 updated 2026-10-24
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-kegg-disease-drug -a claude-codeWhat tooluniverse-kegg-disease-drug does
Retrieves disease genes with KEGG_get_disease_genes, drug targets with KEGG_get_drug_targets, connects variants to pathways and builds disease-gene-drug networks. Separates direct drug-gene relationships (the drug binds the gene product) from indirect ones (the drug acts on a pathway that includes the gene). Requires Python and the ToolUniverse library.
Before querying, the agent converts external IDs to KEGG namespace via KEGG_convert_ids (works KEGG to external only; reverse direction is manual). For human genes, KEGG IDs are hsa: plus the NCBI Gene ID. Absence from KEGG does not imply lack of biological relevance — KEGG coverage is incomplete.
What it helps with
- find genes linked to a disease in KEGG
- build disease-gene-drug networks for drug repurposing
- distinguish direct drug targets from indirect pathway co-membership
- link variants to pathways and diseases
Example prompts for the agent
- Find genes linked to breast cancer in KEGG
- Show vemurafenib targets from KEGG Drug
- Build a KEGG network for melanoma and BRAF variants
How to install tooluniverse-kegg-disease-drug
1Skills CLI (Vercel)
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-kegg-disease-drug -a claude-code2Manually: put the files in place
~/.claude/skills/tooluniverse-kegg-disease-drug/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/tooluniverse-kegg-disease-drug/
1Skills CLI (Vercel)
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-kegg-disease-drug -a codex2Manually: put the files in place
.agents/skills/tooluniverse-kegg-disease-drug/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-kegg-disease-drug -a cursor2Manually: put the files in place
.cursor/skills/tooluniverse-kegg-disease-drug/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install mims-harvard/ToolUniverse skills/tooluniverse-kegg-disease-drug2Manually: put the files in place
.github/skills/tooluniverse-kegg-disease-drug/SKILL.md1Skills CLI (Vercel)
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-kegg-disease-drug -a gemini-cli2Platform CLI
gemini skills install https://github.com/mims-harvard/ToolUniverse --path skills/tooluniverse-kegg-disease-drugManually: put the files in place
.devin/skills/tooluniverse-kegg-disease-drug/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add mims-harvard/ToolUniverse --skill tooluniverse-kegg-disease-drug -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about tooluniverse-kegg-disease-drug
How do I install tooluniverse-kegg-disease-drug in Claude Code?
Run npx skills add mims-harvard/ToolUniverse --skill tooluniverse-kegg-disease-drug -a claude-code and the skill lands in ~/.claude/skills/tooluniverse-kegg-disease-drug/SKILL.md.
What does tooluniverse-kegg-disease-drug do?
It queries KEGG databases to link diseases to genes, drugs to molecular targets and variants to pathways. Builds disease-gene-drug networks and distinguishes direct drug-target binding from indirect pathway co-membership.
What is the difference between direct and indirect drug-gene relationships?
Direct means the drug binds the gene product — a confirmed molecular target. Indirect means the drug affects a pathway that includes the gene; these relationships are weaker evidence.
Source text
SKILL.md · Apache-2.0
---
name: tooluniverse-kegg-disease-drug
description: KEGG-based disease-drug-variant network research. Connects diseases to causal genes, drugs to molecular targets, and variants to pathways using KEGG's editorially curated databases (KEGG Disease, Drug, Network, Variant, Pathway). Use for drug repurposing via shared pathways, mechanistic disease-gene-drug networks, and pathway-based target discovery. Distinguishes direct (binding) vs indirect (pathway co-membership) drug-target relationships.
triggers:
- keywords: [KEGG disease, KEGG drug, KEGG network, KEGG variant, disease genes, drug targets]
- patterns: ["KEGG .* disease", "KEGG .* drug", "disease.*drug.*variant", "KEGG network"]
disable-model-invocation: true
---
# KEGG Disease-Drug-Variant Research
Systematic exploration of disease-drug-variant relationships using KEGG's curated databases.
## Reasoning Strategy
KEGG maps diseases to pathways and drugs to targets, but the real value is in the connections — which pathways link a disease gene to a drug target? This is a network question, not a simple lookup. A gene appearing in a KEGG disease entry has been editorially reviewed as mechanistically relevant; a drug entry with a confirmed target is more reliable than one inferred from pathway co-membership. When using KEGG for drug repurposing, always ask: is the drug-target relationship direct (the drug binds the gene product) or indirect (the drug affects a pathway that contains the gene)? Direct relationships are far stronger evidence. KEGG coverage is not exhaustive — absence from KEGG does not mean absence of biological involvement; complement with Reactome, WikiPathways, or CTD for broader coverage. ID namespace differences are a frequent source of errors: KEGG uses its own gene IDs (e.g., hsa:7157 for TP53), so always convert external IDs before querying KEGG-specific tools.
**LOOK UP DON'T GUESS**: Do not assume KEGG disease IDs, drug IDs, or gene IDs from memory — always search first with `KEGG_search_disease`, `KEGG_search_drug`, or `KEGG_convert_ids`. Do not assume which pathways link a disease gene to a drug; use `KEGG_link_entries` and `KEGG_get_network` to retrieve the actual connections.
## When to Use
- "What genes are associated with [disease] in KEGG?"
- "Find KEGG drugs targeting [gene/pathway]"
- "What variants are linked to [disease] in KEGG?"
- "Show the KEGG disease-gene-drug network for [condition]"
- "Find drugs targeting BRAF variants in cancer"
## Tool Inventory (12 tools)
| Tool | Key Params | Returns |
|------|-----------|---------|
| KEGG_search_disease | `keyword` | Disease entries matching keyword |
| KEGG_get_disease | `disease_id` (e.g., "H00004") | Disease details: genes, drugs, pathways |
| KEGG_get_disease_genes | `disease_id` | All genes for a disease |
| KEGG_search_drug | `keyword` | Drug entries matching keyword |
| KEGG_get_drug | `drug_id` (e.g., "D00123") | Drug details: targets, pathways, metabolism |
| KEGG_get_drug_targets | `drug_id` | Molecular targets for a drug |
| KEGG_search_network | `keyword` | Network entries (disease-gene-drug) |
| KEGG_get_network | `network_id` | Network details and relationships |
| KEGG_search_variant | `keyword` | Variant entries matching keyword |
| KEGG_get_variant | `variant_id` | Variant details and disease associations |
| KEGG_convert_ids | `kegg_id` (IDs joined with `+`), `target_db` | Convert KEGG gene/compound IDs to external IDs (`uniprot`, `ncbi-geneid`, `ncbi-proteinid`, `chebi`, `pubchem`); KEGG → external only |
| KEGG_link_entries | `source`, `target` | Find cross-database relationships (e.g., all genes linked to a pathway, all drugs linked to a disease) |
## Workflow
```
Phase 1: Disease Lookup -> Phase 2: Disease Genes -> Phase 3: Drug Search
-> Phase 4: Drug Targets -> Phase 5: Network/Variant Context -> Report
```
### Phase 1: Disease Lookup
Search and retrieve KEGG disease entries.
```python
# Search for cancer-related diseases
diseases = tu.tools.KEGG_search_disease(keyword="breast cancer")
# Get details for a specific disease
disease = tu.tools.KEGG_get_disease(disease_id="H00031")
```
### Phase 2: Disease Genes
Get genes associated with a KEGG disease entry.
```python
genes = tu.tools.KEGG_get_disease_genes(disease_id="H00031")
```
### Phase 3: Drug Search
Find KEGG drugs by name, target, or keyword.
```python
drugs = tu.tools.KEGG_search_drug(keyword="vemurafenib")
drug_detail = tu.tools.KEGG_get_drug(drug_id="D09996")
```
### Phase 4: Drug Targets
Get molecular targets for a drug.
```python
targets = tu.tools.KEGG_get_drug_targets(drug_id="D09996")
```
### Phase 5: Network & Variant Context
Explore disease-gene-drug networks and variant annotations.
```python
# Search networks linking disease, genes, and drugs
networks = tu.tools.KEGG_search_network(keyword="BRAF melanoma")
network = tu.tools.KEGG_get_network(network_id="N00001")
# Search and get variant details
variants = tu.tools.KEGG_search_variant(keyword="BRAF V600E")
variant = tu.tools.KEGG_get_variant(variant_id="hsa:BRAF")
```
## Example Workflow: Find Drugs Targeting BRAF Variants in Cancer
```python
from tooluniverse import ToolUniverse
tu = ToolUniverse()
tu.load_tools()
# 1. Find BRAF-related diseases
diseases = tu.tools.KEGG_search_disease(keyword="BRAF")
# 2. Get disease genes for melanoma
genes = tu.tools.KEGG_get_disease_genes(disease_id="H00038")
# 3. Search for BRAF-targeting drugs
drugs = tu.tools.KEGG_search_drug(keyword="BRAF inhibitor")
# 4. Get targets for vemurafenib
targets = tu.tools.KEGG_get_drug_targets(drug_id="D09996")
# 5. Get BRAF variant info
variants = tu.tools.KEGG_search_variant(keyword="BRAF V600E")
# 6. Explore disease-gene-drug network
networks = tu.tools.KEGG_search_network(keyword="BRAF melanoma")
```
## ID Conversion & Cross-Linking
Use `KEGG_convert_ids` to map KEGG identifiers to external databases (it converts KEGG → external only):
```python
# Convert KEGG human gene IDs to UniProt accessions (join several IDs with '+')
result = tu.tools.KEGG_convert_ids(kegg_id="hsa:672+hsa:675", target_db="uniprot")
# result['data'] -> [{"kegg_id": "hsa:672", "external_id": "up:P38398"}, ...]
# Convert a KEGG gene ID to its NCBI Gene ID
result = tu.tools.KEGG_convert_ids(kegg_id="hsa:672", target_db="ncbi-geneid")
```
For the reverse direction, KEGG human gene IDs are simply `hsa:` plus the NCBI Gene ID (NCBI Gene 672 is `hsa:672`); there is no external → KEGG conversion tool.
Use `KEGG_link_entries` to retrieve relationships between KEGG databases:
```python
# Find all KEGG pathway IDs that contain a given gene
result = tu.tools.KEGG_link_entries(source="hsa:7157", target="pathway")
# Find all genes linked to a specific pathway
result = tu.tools.KEGG_link_entries(source="hsa05210", target="hsa")
```
These tools are especially useful when you have external IDs (Entrez Gene, UniProt, ChEMBL) and need to bridge into KEGG's namespace, or when you want a complete gene-pathway or drug-disease adjacency list.
## Integration with Other Skills
- **Pathway details**: Use `tooluniverse-systems-biology` for Reactome/WikiPathways cross-ref
- **Drug mechanisms**: Use `tooluniverse-drug-mechanism-research` for ChEMBL/DailyMed MOA
- **Clinical variants**: Use `tooluniverse-cancer-variant-interpretation` for CIViC/ClinVar
- **Drug safety**: Use `tooluniverse-adverse-event-detection` for FAERS data
## Reasoning Framework for Result Interpretation
### Evidence Grading
| Grade | Criteria | Example |
|-------|----------|---------|
| **Strong** | KEGG disease entry with curated gene list, drug with confirmed target, pathway mechanistically linked | H00031 (breast cancer) with BRCA1/BRCA2 genes, D09996 (vemurafenib) targeting BRAF |
| **Moderate** | Disease-gene link in KEGG but no drug-target validation, or network entry without variant data | KEGG disease entry lists gene, but drug targets are inferred from pathway membership |
| **Weak** | Keyword search hit only, no curated disease-gene-drug relationship in KEGG | Drug found by name search but not linked to the disease in KEGG network |
| **Insufficient** | No KEGG entries found, or only cross-database ID conversion available | Rare disease not curated in KEGG Disease |
### Interpretation Guidance
- **KEGG pathway significance**: KEGG pathways are manually curated maps of molecular interactions. A gene appearing in a KEGG disease pathway has been editorially reviewed as relevant to that disease mechanism. However, KEGG coverage is not exhaustive -- absence from KEGG does not mean absence of involvement. Cross-reference with Reactome or WikiPathways for broader coverage.
- **Disease-drug network interpretation**: KEGG Network entries (N-codes) link diseases, genes, and drugs in mechanistic triangles. A drug targeting a gene in a disease network has a curated rationale for therapeutic relevance. The network structure distinguishes direct targets (drug binds gene product) from pathway-level connections (drug affects pathway containing the gene). Prioritize direct target relationships for drug repurposing hypotheses.
- **Variant impact assessment**: KEGG Variant entries are curated for clinical significance (often cancer driver mutations). A variant listed in KEGG with a linked drug entry indicates an established pharmacogenomic or precision oncology relationship (e.g., BRAF V600E linked to vemurafenib). Variants not in KEGG may still be clinically relevant -- cross-reference with ClinVar and CIViC.
- **ID conversion caveat**: KEGG uses its own gene ID namespace (e.g., hsa:7157 for TP53). Human KEGG gene IDs are `hsa:` plus the NCBI Gene ID; `KEGG_convert_ids` only maps KEGG IDs to external IDs (UniProt, NCBI Gene), so resolve UniProt accessions to NCBI Gene IDs first with another tool. Failed conversions may indicate the gene is not in KEGG's curated set.
- **Drug entry completeness**: KEGG Drug entries vary in detail. Approved drugs typically have full target, pathway, and metabolism information. Investigational compounds may have partial entries. Check the drug's "Target" and "Pathway" fields for completeness before drawing conclusions.
### Synthesis Questions
1. Does the KEGG disease entry list the gene of interest with a direct mechanistic role, or is the gene only peripherally connected through a shared pathway?
2. For drug-target relationships, is the target confirmed by KEGG Network (direct link), or inferred from pathway co-membership?
3. Are there KEGG variant entries linking specific mutations to drug response, supporting precision medicine applications?
4. Does the KEGG disease-gene-drug network for the condition align with evidence from other curated sources (CIViC, OncoKB, DrugBank)?
5. If KEGG has limited entries for the query, which complementary databases (Reactome, WikiPathways, CTD) should be consulted to fill gaps?
---
## Output
Markdown report with:
1. Disease summary (KEGG ID, name, associated genes/pathways)
2. Gene list with KEGG gene IDs and symbols
3. Drug candidates with targets and mechanisms
4. Network relationships (disease-gene-drug triangles)
5. Variant annotations if available
© mims-harvard · Apache-2.0 · mims-harvard/ToolUniverse