chem-tools

Featurize molecules and predict binding affinity for drug discovery workflows.

126|8|Updated Mar 10, 2026
One-click install
npx skills add https://github.com/DrugClaw/DrugClaw --skill chem-tools
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: chem-tools
Source: https://github.com/DrugClaw/DrugClaw/tree/main/skills/pharma/chem-tools
Command: npx skills add https://github.com/DrugClaw/DrugClaw --skill chem-tools

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires deepchem, pyscf, rdkit, numpy, pandas, sklearn, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill streamlines complex computational chemistry tasks, enabling faster analysis and prediction in drug discovery workflows.

Core Features & Use Cases

  • Molecule Featurization: Generate molecular fingerprints and descriptors using DeepChem and RDKit.
  • ADMET Triage: Quickly assess drug-likeness and identify potential liabilities.
  • QSAR & Affinity Prediction: Train and apply models for binding affinity and bioactivity prediction.
  • Structure-Based Analysis: Prepare structure datasets and perform protein-ligand affinity benchmarking.
  • DrugBank Lookup: Search DrugBank for drug information and structures.
  • QM Calculations: Run single-point quantum mechanical calculations with PySCF.
  • Use Case: Predict the binding affinity of a new set of drug candidates to a target protein using structure-aware models, followed by a virtual screening ranking that integrates multiple predictive signals.

Quick Start

Use the chem-tools skill to featurize molecules from SMILES strings using RDKit Morgan fingerprints.

Frequently Asked Questions about chem-tools

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

FAQPage Schema
How do I predict binding affinity for new drug candidates using SMILES strings?

To predict binding affinity for drug candidates, you can featurize molecules from SMILES strings using RDKit Morgan fingerprints, then apply QSAR and structure-aware models to estimate bioactivity and rank compounds.

What's the best way to run ADMET screening for drug-likeness and potential liabilities?

ADMET triage assesses drug-likeness by screening molecules to quickly identify potential absorption, distribution, metabolism, excretion, and toxicity liabilities before committing to extensive experimental testing.

Can I use DeepChem with RDKit for molecular featurization in computational chemistry?

Yes, DeepChem and RDKit are supported together to generate molecular fingerprints and descriptors, allowing you to featurize molecules from SMILES strings for downstream QSAR modeling and virtual screening.

Does this computational chemistry workflow support quantum mechanical calculations?

Quantum mechanical calculations are supported through PySCF, enabling you to run single-point QM calculations to analyze electronic properties for molecules in your drug discovery pipeline.

How do I perform protein-ligand affinity benchmarking with structure-based analysis?

Structure-based analysis prepares structure datasets and integrates docking results to perform protein-ligand affinity benchmarking, generating a virtual screening ranking that combines multiple predictive signals.

How do I look up drug information and structures in DrugBank?

DrugBank lookup searches a database to retrieve drug information and structures, providing a reference point for computational chemistry workflows and target identification during drug discovery.