
compound-profile
by huifer
Drug Discovery Intelligence plugin for Claude Code. AI-powered target validation, competitive intelligence, literature analysis & clinical trials insights. Integrates Open Targets, ChEMBL, PubMed & 5+ databases.
SKILL.md
name: compound-profile description: | Generate comprehensive compound profiles including structure, properties, bioactivity, and development status. Use for drug analysis, SAR studies, and competitive profiling.
Keywords: compound, drug, molecule, structure, SMILES, bioactivity, IC50 category: Compound Analysis tags: [compound, drug, structure, bioactivity, chembl] version: 1.0.0 author: Drug Discovery Team dependencies:
- chembl-database
- pubchem-database
- drugbank-database
Compound Profile Skill
Comprehensive compound analysis for drug discovery and medicinal chemistry.
Quick Start
/compound erlotinib
/compound-profile CC1=CC=C(C=C1)CNC(=O)C1=NC=C(C=C1)N
Analyze osimertinib properties and bioactivity
Compare gefitinib, erlotinib, afatinib profiles
What's Included
| Section | Description | Data Source |
|---|---|---|
| Basic Info | Name, type, status, company | ChEMBL, DrugBank |
| Structure | SMILES, InChI, molecular weight | PubChem, ChEMBL |
| Properties | LogP, HBD, HBA, TPSA, RO5 | Calculated, PubChem |
| Bioactivity | Target affinity, IC50, Ki | ChEMBL, BindingDB |
| Development | Phase, indications, status | Drugs@FDA |
| Similar Compounds | Structure similarity search | ChEMBL |
| Safety | Known toxicity, warnings | SIDER, PubChem |
Output Structure
# Compound Profile: Erlotinib
## Executive Summary
Erlotinib is a first-generation EGFR TKI approved for NSCLC (2004).
Key characteristics: Oral bioavailability, good brain penetration,
resistance mutations limit long-term efficacy.
## Basic Information
| Field | Value |
|-------|-------|
| Name | Erlotinib |
| Brand Names | Tarceva |
| ChEMBL ID | CHEMBL880 |
| Type | Small molecule |
| Class | Kinase inhibitor |
| Status | Approved |
| Approval Year | 2004 |
| Company | Astellas (OSI) |
| Indications | NSCLC, pancreatic cancer |
## Structure & Properties
**SMILES:** `COc1cc2nc(Nc3ccc(Oc4ccc(O)cc4)cc3)nc2cc1OC`
| Property | Value | Rule of 5 Check |
|----------|-------|----------------|
| MW | 393.4 Da | ✓ (<500) |
| LogP | 3.1 | ✓ (<5) |
| HBD | 1 | ✓ (≤5) |
| HBA | 7 | ✓ (≤10) |
| TPSA | 76.3 Ų | ✓ (<140) |
| Rotatable Bonds | 6 | |
## Bioactivity
| Target | Type | Affinity | Units |
|--------|------|----------|-------|
| EGFR | IC50 | 0.5 | nM |
| ERBB2 | IC50 | 1200 | nM |
| LCK | IC50 | 5 | nM |
## Development History
| Year | Milestone |
|------|-----------|
| 2004 | FDA Approval (NSCLC) |
| 2005 | EMEA Approval |
| 2010 | Pancreatic cancer approval |
| 2011 | Generic launch (US) |
## Similar Compounds
| Compound | Similarity | Difference |
|----------|------------|------------|
| Gefitinib | 85% | Different core scaffold |
| Afatinib | 72% | Irreversible binder |
| Osimertinib | 68% | 3rd-gen, mutant-selective |
| Icotinib | 82% | China-approved analog |
## Safety Profile
**Common AEs:** Rash, diarrhea, fatigue, anorexia
**Boxed Warning:** Interstitial lung disease
**Contraindications:** Hypersensitivity to erlotinib
## Patent Status
| Patent | Number | Expiry |
|--------|---------|--------|
| Base patent | US5747498 | 2019 (expired) |
| Formulation | US6943129 | 2020 |
| Method of use | US6900221 | 2021 |
Examples
By Name
/compound erlotinib
/compound-profile sotorasib
By Structure
/compound "CC1=CC=C(C=C1)CNC(=O)C1=NC=C(C=C1)N"
/compound-profile SMILES
Comparison
Compare compounds erlotinib, gefitinib, afatinib
Analyze bioactivity across EGFR inhibitors
Property Analysis
/compound erlotinib --focus properties
Analyze drug-likeness of this compound
Check Lipinski rule of 5 violations
Running Scripts
# Fetch compound by name
python scripts/fetch_compound_data.py erlotinib --output compound.json
# Fetch by SMILES
python scripts/fetch_compound_data.py --smiles "CC1=CC=C..." -o data.json
# Similarity search
python scripts/fetch_compound_data.py --similar CHEMBL880 --threshold 0.7
# Bioactivity summary
python scripts/fetch_compound_data.py erlotinib --bioactivity -o activity.json
# Structure search
python scripts/fetch_compound_data.py --scaffold quinazoline --limit 20
Requirements
pip install requests pandas rdkit
Additional Resources
Best Practices
- Use standard names: Generic names preferred over brand
- Verify ChEMBL ID: Most reliable identifier
- Check bioactivity: Cross-reference multiple sources
- Analog analysis: Use similarity searches for SAR
- Validate SMILES: Check structure validity
Common Pitfalls
| Pitfall | Solution |
|---|---|
| Name ambiguity | Use ChEMBL ID when possible |
| Stereochemistry | SMILES may not capture isomerism |
| Outdated data | Check multiple sources |
| Salt forms | API may have multiple entries |
| Tautomerism | Different SMILES for same structure |
Score
Total Score
Based on repository quality metrics
SKILL.mdファイルが含まれている
ライセンスが設定されている
100文字以上の説明がある
GitHub Stars 100以上
1ヶ月以内に更新
10回以上フォークされている
オープンIssueが50未満
プログラミング言語が設定されている
1つ以上のタグが設定されている
Reviews
Reviews coming soon
