skills.rest
xu1713
xu1713COMMUNITY
Explore all skills by this author →
GitHub

drug-discovery

Query ChEMBL, PubChem, OpenFDA, and OpenTargets APIs for compound properties and drug-likeness analysis.

Updated Jun 5, 2026
Search Tags:#drug discovery#medicinal chemistry#cheminformatics#pubchem#chembl#admet#lipinski
One-click install
npx skills add https://github.com/xu1713/openhorse --skill drug-discovery-xu1713
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: drug-discovery
Source: https://github.com/xu1713/openhorse/tree/main/openhorse/openhorse/optional-skills/research/drug-discovery
Command: npx skills add https://github.com/xu1713/openhorse --skill drug-discovery-xu1713

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve? Pharmaceutical research requires querying multiple public databases and applying medicinal chemistry rules manually. This Skill automates bioactive compound searches, drug-likeness screening, drug interaction lookups, and ADMET analysis using free public APIs. ## Core Features & Use Cases - Bioactive Compound Search: Query ChEMBL by target name or ID to retrieve bioactivity data (pChEMBL values, IC50, Ki) for active molecules. - Drug-Likeness Screening: Calculate Lipinski Rule of Five, Veber rules, TPSA, and QED scores for any compound via PubChem, with batch screening support. - Safety & Interaction Lookup: Retrieve drug-drug interactions from OpenFDA labels and top reported adverse events from FDA event data. - Use Case: A medicinal chemist evaluating a lead series can batch-screen 20 candidate molecules against Ro5 and Veber rules, then check the top hit for CYP450 interaction liabilities and hERG risk using the ADMET reference guide. ## Quick Start Ask the assistant to check whether aspirin passes the Lipinski Rule of Five and look up its known drug interactions.

Frequently Asked Questions about drug-discovery

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check if a compound passes the Lipinski Rule of Five?▼

Provide the compound name and the Skill fetches molecular weight, LogP, H-bond donors, and acceptors from the PubChem API, then evaluates all four Ro5 criteria plus Veber rules. The ro5_screen.py script supports batch screening of multiple compounds at once.

How do I find bioactive compounds for a protein target like EGFR?▼

The Skill searches the ChEMBL target API by name, retrieves the target's ChEMBL ID, then queries activity data filtered by pChEMBL threshold. The chembl_target.py script returns a ranked table of active molecules with potency values.

Do ChEMBL, PubChem, and OpenFDA APIs require API keys?▼

No, all APIs used are free and public with no authentication required. ChEMBL recommends adding a one-second delay between batch requests to respect rate limits.

Can I look up drug-drug interactions without a paid database?▼

Yes, the Skill queries the OpenFDA drug label endpoint for interaction sections and the drug event endpoint for top reported adverse reactions. Note that FDA adverse event data reflects reports, not proven causation.

What are the limitations of API-based drug-likeness screening?▼

Screening relies on PubChem's precomputed properties rather than local RDKit calculation, so novel or unindexed structures cannot be evaluated by name. Predictions are rule-based estimates and do not replace experimental ADMET assays or clinical judgment.

Related Skills

Explore More Education & Research Skills→
InternScienceInternScience
28

molclaw-admet

Predicts ADMET properties for batches of molecules from SMILES strings or files.

Official
Intermediate
InternScienceInternScience
28

molclaw-pulchura-rebuild

Rebuild missing atoms in protein PDB structures for docking and simulation preparation.

Official
Intermediate
learningmatter-mitlearningmatter-mit
144

general-chemical-literature

Retrieve PubMed and patent cross-references for a chemical compound via PubChem CID.

Official
Basic
K-Dense-AIK-Dense-AI
41.1k

molfeat

Convert SMILES molecules into ML-ready features using 100+ featurizers and pretrained models.

Official
Intermediate