molfeat

Convert SMILES strings into over 100 molecular fingerprints and descriptors.

1|Updated Jan 14, 2026

Search Tags:#machine learning #deep learning #cheminformatics #fingerprints #descriptors #molecular featurization

Ready to run

npx skills add https://github.com/Sologa/codex-pipeline --skill molfeat-sologa

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: molfeat
Source: https://github.com/Sologa/codex-pipeline/tree/main/.codex/skills/molfeat
Command: npx skills add https://github.com/Sologa/codex-pipeline --skill molfeat-sologa

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes 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 models.

Core Features & Use Cases

100+ Featurizers: Access a vast library of fingerprints, descriptors, and deep learning embeddings.
SMILES to Features: Effortlessly convert SMILES strings into ML-ready vectors.
Use Case: You need to build a QSAR model to predict drug efficacy. Use this Skill to generate ECFP fingerprints for your dataset of drug molecules, preparing them for input into a machine learning classifier.

Quick Start

Use the molfeat skill to generate ECFP fingerprints for the molecule with SMILES 'CCO'.

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 into ECFP fingerprints for machine learning?▼

To convert SMILES strings into ECFP fingerprints for machine learning, you provide the SMILES text to a molecular featurization tool. This Skill processes the chemical structures and outputs numerical vectors ready for ML model input.

What molecular descriptors are available for QSAR modeling and virtual screening?▼

For QSAR modeling and virtual screening, over 100 featurizers are available including ECFP, MACCS, and RDKit descriptors. These generate numerical representations of chemical structures required for machine learning algorithms.

Can I use pre-trained embeddings like ChemBERTa for molecular featurization?▼

Yes, pre-trained embeddings like ChemBERTa are supported for molecular featurization. The library provides transformer-based models that convert SMILES strings and RDKit molecules into deep learning embeddings for ML applications.

Does molecular featurization work with RDKit molecule objects as input?▼

Yes, molecular featurization works with RDKit molecule objects as input. The library supports converting both SMILES strings and RDKit molecules into over 100 types of fingerprints, descriptors, and pre-trained embeddings.

What is the best way to prepare chemical structures for a machine learning classifier?▼

The best way to prepare chemical structures for a machine learning classifier is to apply molecular featurization. Converting molecular structures into numerical vectors like ECFP fingerprints or RDKit descriptors generates the ML-ready input required for model training.