rdkit

Computes cheminformatics features and transformations from molecular structures using RDKit APIs.

4|Updated Mar 2, 2026
One-click install
npx skills add https://github.com/shushuzn/Rairos --skill rdkit-shushuzn
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rdkit
Source: https://github.com/shushuzn/Rairos/tree/main/skills/rdkit
Command: npx skills add https://github.com/shushuzn/Rairos --skill rdkit-shushuzn

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

RDKit helps you turn chemical structures into computed descriptors, fingerprints, similarities, and queryable substructures so you can screen, filter, and investigate molecules without manual, error-prone chemistry tooling.

Core Features & Use Cases

  • Molecular I/O (SMILES/SDF/MOL/InChI): Read and write structures in common cheminformatics formats, then validate parsing and sanitization results.
  • Descriptors and drug-likeness metrics: Compute physicochemical and structural properties like MW, LogP, TPSA, H-bond donors/acceptors, ring counts, and Lipinski/QED-style signals.
  • Fingerprints, similarity, and clustering: Generate RDKit/Morgan/atom-pair/torsion fingerprints, compute Tanimoto/Dice/Cosine similarities, and cluster by fingerprint diversity.
  • SMARTS substructure search and reaction handling: Build SMARTS queries for inclusion/exclusion screening and run reaction SMARTS to generate products.
  • 2D/3D coordinate generation and depiction: Compute 2D coordinates for diagrams, generate conformers, optimize geometry, and visualize highlighted environments.
  • Use Case: Given a hit set and a target scaffold, compute descriptors and fingerprints, screen for substructure matches using SMARTS, rank candidates by similarity above a threshold, and export results for downstream research workflows.

Quick Start

Use the rdkit skill to fingerprint a query SMILES and screen a molecule library for matches above a similarity threshold.

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 from SMILES for drug-likeness screening?

To calculate molecular descriptors from SMILES, parse the structure and compute physicochemical properties like MW, LogP, TPSA, and H-bond donors/acceptors. You can then evaluate drug-likeness signals using Lipinski or QED-style metrics for screening.

How do I run a SMARTS substructure search on a molecule library?

Running a SMARTS substructure search involves building SMARTS queries against parsed molecular structures to identify inclusion or exclusion matches. This screens the molecule library for specific substructural features efficiently.

What is the best way to compute Tanimoto similarity between molecular fingerprints?

Computing Tanimoto similarity requires generating RDKit, Morgan, atom-pair, or torsion fingerprints for the molecules, then applying the similarity metric with a configurable threshold to rank candidates by structural diversity.

Can I parse and export multiple chemical structure formats like SDF and InChI?

Yes, you can parse and export multiple chemical structure formats including SMILES, SDF, MOL, and InChI. The process validates parsing and sanitization results to ensure structural integrity before exporting.

How do I apply reaction SMARTS to generate chemical products?

Applying reaction SMARTS involves defining reaction patterns and running them against reactant structures to automatically generate the corresponding chemical products for downstream synthesis workflows.

Does this approach support generating 2D and 3D coordinates for molecular visualization?

Yes, this approach supports computing 2D coordinates for structural diagrams and generating 3D conformers. It optimizes molecular geometry and visualizes highlighted environments for detailed analysis.