molfeat

Convert SMILES strings and RDKit molecules into numerical feature vectors.

8|Updated Jan 13, 2026

Search Tags:#machine learning #drug discovery #cheminformatics #chemistry #featurization #molecular

Ready to run

npx skills add https://github.com/hxk622/TokenDance --skill molfeat-hxk622

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: molfeat
Source: https://github.com/hxk622/TokenDance/tree/main/backend/app/skills/builtin/scientific/chemistry/molfeat
Command: npx skills add https://github.com/hxk622/TokenDance --skill molfeat-hxk622

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill simplifies the complex process of converting molecular structures into numerical representations required for machine learning, enabling faster and more accurate cheminformatics and drug discovery.

Core Features & Use Cases

100+ Featurizers: Access a vast library of pre-trained embeddings and hand-crafted molecular descriptors.
SMILES to Features: Effortlessly convert SMILES strings or RDKit molecules into ML-ready feature vectors.
Use Case: Predict the properties of new drug candidates by featurizing their molecular structures using ECFP fingerprints and feeding them into a QSAR model.

Quick Start

Use the molfeat skill to convert the SMILES string 'CCO' into ECFP features.

Frequently Asked Questions about molfeat

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

FAQPage Schema

How do I convert SMILES strings to ECFP fingerprints for machine learning?▼

You can convert SMILES strings to ECFP fingerprints by applying molecular featurization, which transforms chemical structures into ML-ready numerical vectors. This supports SMILES and RDKit molecules for QSAR modeling and virtual screening.

What is molecular featurization for drug discovery?▼

Molecular featurization for drug discovery is the process of converting molecular structures into numerical representations required for machine learning. It generates hand-crafted descriptors or pre-trained deep learning embeddings for cheminformatics analysis.

Can I use pre-trained embeddings like ChemBERTa for QSAR modeling?▼

Yes, you can use pre-trained embeddings like ChemBERTa for QSAR modeling. Molecular featurization provides over 100 featurizers, including pre-trained deep learning embeddings like ChemBERTa and GIN, to predict drug candidate properties.

Does molecular featurization support RDKit descriptors and Mordred?▼

Yes, molecular featurization supports RDKit descriptors and Mordred. It offers a library of over 100 featurizers including hand-crafted molecular descriptors like RDKit and Mordred alongside fingerprints such as ECFP and MACCS.

What's the best way to generate numerical representations for virtual screening?▼

The best way to generate numerical representations for virtual screening is using molecular featurization to convert SMILES strings into ML-ready feature vectors. This facilitates similarity searching by providing diverse featurizers like ECFP fingerprints.

Do I need RDKit molecules to perform molecular featurization?▼

No, you do not need RDKit molecules to perform molecular featurization, as it directly accepts SMILES strings as input. However, it also fully supports RDKit molecules for conversion into over 100 types of ML-ready feature vectors.