rdkit

Parse molecular inputs and compute descriptors, fingerprints, and SMARTS-based similarity searches.

21|2|Updated Dec 8, 2025
One-click install
npx skills add https://github.com/silverstein/claude-scientific-skills-desktop --skill rdkit-silverstein
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rdkit
Source: https://github.com/silverstein/claude-scientific-skills-desktop/tree/main/corpus/rdkit
Command: npx skills add https://github.com/silverstein/claude-scientific-skills-desktop --skill rdkit-silverstein

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill helps you parse chemical structures and turn them into computed properties, fingerprints, similarity hits, and actionable substructure results—without manually stitching together RDKit workflows.

Core Features & Use Cases

  • Molecular I/O & Structure Handling: Read/write molecules from SMILES, SDF/MOL, InChI, and generate 2D/3D coordinates for downstream analysis.
  • Descriptors & Drug-Likeness Signals: Compute physicochemical properties (MW, LogP, TPSA, HBD/HBA, rotatable bonds) and common scoring like QED and Lipinski/lead-like checks.
  • Similarity & Substructure Search: Generate multiple fingerprint types, compute Tanimoto/Dice/Cosine similarity, and run SMARTS-based substructure filtering.
  • Reactions & Visualization: Define reaction SMARTS, run reactions, and draw molecules with highlighted atom environments for interpretation.

Quick Start

Use the rdkit skill to compute drug-likeness descriptors, run a similarity screening against a set of SMILES, and highlight substructure matches using SMARTS in one workflow.

Frequently Asked Questions about rdkit

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

FAQPage Schema
How do I calculate molecular descriptors and evaluate drug-likeness from SMILES?

Molecular descriptor calculation parses SMILES to compute physicochemical properties like MW, LogP, and TPSA, then applies Lipinski or QED scoring to evaluate drug-likeness for screening.

How does SMARTS-based substructure search work for molecular screening?

SMARTS substructure search works by matching query patterns against parsed molecules to filter batches, returning interpretable results that highlight matched atom environments for scaffold and feature exploration.

Can I run Tanimoto similarity search across a batch of molecules using fingerprints?

Yes, you can compute Tanimoto, Dice, or Cosine similarity metrics across batches of molecules by generating configurable fingerprints to screen and cluster similar chemical structures.

What chemical file formats are supported for molecule I/O and parsing?

Supported formats for molecule I/O include SMILES, SDF/MOL, and InChI, which are parsed with sanitization-aware logic to generate 2D or 3D coordinates for downstream cheminformatics analysis.

Do I need to manually configure RDKit workflows to highlight substructure matches?

No, you can run SMARTS query matching and draw molecules with highlighted atom environments for interpretation without manually stitching together complex RDKit visualization workflows.

Can I define and run chemical reactions using SMARTS patterns?

Yes, you can define reaction SMARTS patterns, execute reactions on parsed inputs, and visualize products to support structure-activity relationship analysis in drug discovery workflows.