skills.rest
autonomous-ai
autonomous-aiOFFICIAL
Explore all skills by this author →
Website𝕏@autonomous_labsGitHub

rdkit

Build molecules from SMILES with RDKit and compute properties, conformers, and similarity.

465|41|Updated Aug 4, 2026
Search Tags:#cheminformatics#similarity search#molecule design#drug-likeness#smiles#conformers#rdkit
One-click install
npx skills add https://github.com/autonomous-ai/openharness --skill rdkit-autonomous-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: rdkit
Source: https://github.com/autonomous-ai/openharness/tree/main/store/agents/rdkit/skills/rdkit
Command: npx skills add https://github.com/autonomous-ai/openharness --skill rdkit-autonomous-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires rdkit, numpy, pandas.

What problem does it solve? Designing and analyzing small molecules requires cheminformatics expertise: parsing SMILES, generating 3D conformers, computing drug-likeness properties, and comparing analogues. This Skill wraps RDKit into a guided workflow that produces molecules, property reports, and conformer ensembles ready for a live 3D viewer. ## Core Features & Use Cases - Molecule building from SMILES or scaffolds: Parse structures, embed 3D conformers with ETKDGv3 and MMFF94 minimization, and write SDF output for visualization. - Property and drug-likeness analysis: Compute MW, cLogP, TPSA, Lipinski/Veber rules, QED, PAINS/Brenk alerts, and per-atom charges. - Analogue series and search: Track parent-child relationships between analogues, run Morgan fingerprint (ECFP4) Tanimoto similarity searches, and match substructures with SMARTS. - Use Case: Design ibuprofen as a lead compound, enumerate halogen and CF3 analogues with property comparison tables, and view each molecule's conformer ensemble and property deltas in the pane. ## Quick Start Ask the agent to design a molecule from a SMILES string, for example: build caffeine from CN1C=NC2=C1C(=O)N(C)C(=O)N2C and show its properties and conformers.

Frequently Asked Questions about rdkit

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

FAQPage Schema
How do I build a molecule from a SMILES string with RDKit?▼

Use the design function with a SMILES string and a name, which parses the molecule, embeds conformers with ETKDGv3, minimizes with MMFF94, and writes SDF and property files. For manual control, use mol_from_smiles followed by embed_conformers and write_outputs.

How do I compute molecular properties like cLogP, TPSA, and QED?▼

Call properties(mol) on any parsed molecule to get MW, cLogP, TPSA, Lipinski and Veber rule checks, QED, Gasteiger charges, and PAINS/Brenk alerts. The design workflow writes these automatically to properties.json and the molecule record.

How do I run a similarity search across a molecule library?▼

Read a .smi file with read_smi, then compute Morgan fingerprint (ECFP4) Tanimoto scores with the similarity function against your query SMILES. Scores above 0.7 indicate close analogues; below 0.3 indicates unrelated structures.

Why does Chem.MolFromSmiles return None for my SMILES?▼

A None result means the SMILES fails sanitization, usually from a typo such as a five-valent carbon, an unclosed ring, or wrong aromaticity. Fix the string itself rather than bypassing sanitization with sanitize=False.

Can RDKit handle stereochemistry and salts in SMILES?▼

Stereochemistry must be written explicitly with [C@H]/[C@@H] or E/Z bond markers, otherwise an unspecified centre embeds as one arbitrary enantiomer. Salts should be stripped with LargestFragmentChooser before computing properties.

What are the limitations of RDKit conformer generation and property scores?▼

Conformer energies are vacuum force-field values that guide shape, not populations, and macrocycles may fail to embed without retries or useMacrocycleTorsions. cLogP, QED, and Lipinski are heuristic estimates and do not predict activity, binding, or safety.

Related Skills

Explore More Education & Research Skills→
mims-harvardmims-harvard
1.7k

tooluniverse-chemical-sourcing

Finds commercial vendors for chemical compounds and compares pricing, purity, and availability.

Official
Advanced
ANakaANaka
3

movie-creation

Automate PyMOL animation and movie creation from structural data.

Community
Advanced
felixboehmfelixboehm

adaptyv

Submit protein sequences for automated experimental assays and computational optimization.

Community
Advanced
pradyumnasagarpradyumnasagar
6

generative-design

Generates novel molecules via de novo design, scaffold decoration, and linker design.

Community
Advanced