QSARmil: A Package for Molecular Multi-Instance Machine Learning
Abstract
Numerous Python tools are available for molecular machine learning, typically built around core modeling paradigms such as descriptor-based approaches (“RDKit + scikit-learn”), deep neural networks, graph neural networks, geometric deep learning, and foundation models. Multi-instance learning (MIL) was originally inspired by the drug discovery problem, where a molecule can adopt multiple conformers and the biologically active one is unknown. Despite its clear relevance, MIL has not yet been supported by a dedicated and comprehensive Python platform for molecular applications. Here, we present QSARmil, a Python package designed for molecular multi-instance machine learning, that provides an integrated workflow including instance generation (e.g., conformers), 2D and 3D descriptor calculation, implementation of diverse MIL algorithms, and key instance identification. By bringing these components together in a unified and user-friendly framework, QSARmil makes it straightforward to apply multi-instance learning to any molecular property prediction task.
Related
- A Machine Learning Approach to logBB Prediction Combining Molecular Docking and Quantum-Chemical Descriptors
- Accuracy, Reliability, and Efficiency of a MACE Machine-Learning Interatomic Potential for Aspirin: A Reproducible Benchmark for Molecular Simulation
- Embedded Morgan Fingerprints for more efficient molecular property predictions with machine learning
- Integration of Network Toxicology and Machine Learning to Prioritize Candidate Molecular Features Associated with NNK-Related Lung Cancer
- The Thiolate Shell as a Dynamic Gold Reservoir: Size-Dependent Thermal Restructuring of Gold Nanoclusters from MACE-Based Machine-Learning Molecular Dynamics
BibTeX
@misc{zankov2026qsarmil,
title = {{QSARmil: A Package for Molecular Multi-Instance Machine Learning}},
author = {Dmitry Zankov and Tomasz Danel and Martin Šícho and Wim Dehaen and Pavel Polishchuk and Mario Barbatti},
howpublished = {ChemRxiv},
year = {2026},
month = sep,
doi = {10.26434/chemrxiv.15008449/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008449/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - QSARmil: A Package for Molecular Multi-Instance Machine Learning
AU - Dmitry Zankov
AU - Tomasz Danel
AU - Martin Šícho
AU - Wim Dehaen
AU - Pavel Polishchuk
AU - Mario Barbatti
PY - 2026
DA - 2026/09/07
PB - ChemRxiv
DO - 10.26434/chemrxiv.15008449/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008449/v1
AB - Numerous Python tools are available for molecular machine learning, typically built around core modeling paradigms such as descriptor-based approaches (“RDKit + scikit-learn”), deep neural networks, graph neural networks, geometric deep learning, and foundation models. Multi-instance learning (MIL) was originally inspired by the drug discovery problem, where a molecule can adopt multiple conformers and the biologically active one is unknown. Despite its clear relevance, MIL has not yet been supported by a dedicated and comprehensive Python platform for molecular applications. Here, we present QSARmil, a Python package designed for molecular multi-instance machine learning, that provides an integrated workflow including instance generation (e.g., conformers), 2D and 3D descriptor calculation, implementation of diverse MIL algorithms, and key instance identification. By bringing these components together in a unified and user-friendly framework, QSARmil makes it straightforward to apply multi-instance learning to any molecular property prediction task.
ER -