molfeat

Convert SMILES or RDKit molecules into ML-ready numerical representations.

52|6|Updated Nov 24, 2025
One-click install
npx skills add https://github.com/ovachiever/droid-tings --skill molfeat-ovachiever
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: molfeat
Source: https://github.com/ovachiever/droid-tings/tree/main/skills/molfeat
Command: npx skills add https://github.com/ovachiever/droid-tings --skill molfeat-ovachiever

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires molfeat, datamol, numpy, scikit-learn, and includes references (resource) components.

What problem does it solve?

Molfeat unifies 100+ featurizers and pretrained transformers to convert SMILES or RDKit molecules into machine-learning-ready representations, enabling rapid model development and comparison.

Core Features & Use Cases

  • Hundreds of featurizers (fingerprints, descriptors, graph embeddings)
  • Transformers and pretrained models for embeddings
  • Batch transformers with scikit-learn compatibility
  • State persistence for reproducible pipelines
  • Supports chemoinformatics workflows, QSAR, and similarity searching

Use cases include building and evaluating ML models on molecular data, docking features, and clustering chemical space.

Quick Start

Featurize a set of SMILES with ECFP fingerprints using a MoleculeTransformer, then downstream ML.

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 numerical features for machine learning?

Molecular featurization converts SMILES or RDKit molecules into ML-ready numerical representations using 100+ featurizers including fingerprints, descriptors, and graph embeddings. Molfeat unifies these featurizers with scikit-learn compatibility and batch processing to enable rapid model development on molecular data.

Can I use ECFP fingerprints and other descriptors in parallel for QSAR modeling?

Yes. Molfeat provides parallelized batch processing across 100+ featurizers including ECFP fingerprints and chemical descriptors, with scikit-learn compatibility for direct integration into QSAR and QSPR workflows.

What's the best way to compare multiple molecular representations for virtual screening?

Molfeat enables rapid comparison by offering hundreds of featurizers—fingerprints, descriptors, and pretrained transformer embeddings—all accessible through a unified interface with batch processing and caching for efficient similarity searching and chemical space analysis.

Does molfeat support pretrained language models like ChemBERTa for molecular embeddings?

Yes. Molfeat includes optional pretrained transformers for generating embeddings, supporting deep learning workflows with state persistence for reproducible pipelines alongside traditional fingerprints and descriptors.

How do I serialize and reproduce molecular featurization pipelines across runs?

Molfeat provides stateful serialization to save and reload featurization pipelines, ensuring reproducible results and enabling consistent feature generation across multiple model development and evaluation cycles.