Unify ML-ready molecular featurization with100+ models and workflows.
npx skills add https://github.com/Hung-3008/agusta --skill molfeat-hung-3008
Please help me install this Agent Skill. Skill: molfeat Source: https://github.com/Hung-3008/agusta/tree/main/.agents/skills/molfeat Command: npx skills add https://github.com/Hung-3008/agusta --skill molfeat-hung-3008
MolFeat provides a unified, scalable platform for converting chemical structures into machine-learning-ready features using 100+ featurizers and pretrained models. It enables researchers to build QSAR models, perform virtual screening, and analyze chemical spaces with both traditional descriptors and modern deep representations.
Launch a basic featurization workflow by creating a FPCalculator and a MoleculeTransformer, featurize a list of SMILES, and optionally save transformer state for reproducibility.
High-intent search queries and answers about installing and using this skill.
Molecular featurization converts SMILES strings into ML-ready features using calculators and transformers. You can apply 100+ featurizers, including fingerprints, descriptors, and graph embeddings, to prepare chemical structures for QSAR modeling and virtual screening.
For QSAR modeling, you can compute diverse molecular descriptors and fingerprints, alongside pretrained language model and graph embeddings. This unified featurization suite provides scikit-learn compatible transformers to seamlessly integrate these representations into your ML pipelines.
Yes, batch molecular featurization is supported through scikit-learn compatible transformers. The modular API provides calculators and pretrained transformers that process lists of SMILES strings, allowing you to save transformer state for reproducible batch workflows.
Chemical space analysis uses molecular embeddings generated from descriptors, fingerprints, or graph models. By unifying these featurizers into a single platform, you can efficiently compute structural representations for similarity searching and scaffold analysis across large compound libraries.
Yes, molecular featurization for virtual screening supports pretrained deep learning models. The centralized model store integrates pretrained transformers, enabling you to apply advanced language model and graph embeddings directly to your chemical structures for ML tasks.
A unified featurization library streamlines chemoinformatics workflows by consolidating 100+ calculators and transformers into one API. This interoperability prevents fragmented codebases, ensuring consistent feature extraction for QSAR, similarity searching, and space analysis.
ovachieverConvert SMILES or RDKit molecules into ML-ready numerical representations.
LeonChaoXConvert SMILES or RDKit molecules into numerical features for machine learning.
jimmc414Convert SMILES strings or RDKit molecules into numerical representations for machine learning.
pur3v4d3rFeaturize SMILES or RDKit molecules into machine-learning-ready vectors with Molfeat.
