19 January 2026 in Digital Discovery
QUED, a QM/ML framework that combines structural and electronic molecular information to build regression models for physicochemical and biological property prediction. Our work highlights the value of QM data for reliable and interpretable models.
18 September 2025 in ChemRxiv
Machine learning (ML) approaches have drastically advanced the exploration of structure-property and property-property relationships in computer-aided drug discovery. A central challenge in this field is the identification of molecular descriptors that can effectively capture both geometric- and electronic structure-derived features, enabling the development of reliable and interpretable predictive models. While numerous descriptors focusing solely on structural characteristics have been recently proposed, improvements in model accuracy often come at the cost of increased computational demands, thereby restricting their practical applicability. To address this challenge, we introduce the “QUantum Electronic Descriptor” (QUED) framework, which integrates both structural and electronic data of molecules to develop ML regression models for property prediction. In doing so, a quantum-mechanical (QM) descriptor is derived from molecular and atomic properties computed using the semi-empirical density functional tight-binding (DFTB) method, which allows for efficient modelling of both small and large drug-like molecules. This descriptor is combined with inexpensive geometric descriptors--capturing two-body and three-body interatomic interactions--to form comprehensive molecular representations used to train Kernel Ridge Regression and XGBoost models. As a proof of concept, we validate QUED using the QM7-X dataset, which comprises equilibrium and non-equilibrium conformations of small drug-like molecules, demonstrating that incorporating electronic structure data notably enhances the accuracy of ML models for predicting physicochemical properties. For biological endpoints, we find that QM properties provide some predictive value for toxicity and lipophilicity prediction, as assessed using the TDCommons-LD50 and the MoleculeNet benchmark datasets. Moreover, a SHapley Additive exPlanations (SHAP) analysis of the toxicity and lipophilicity predictive models reveals that molecular orbital energies and DFTB energy components are among the most influential electronic features. Hence, our work underscores the importance of incorporating QM descriptors to enhance both the accuracy and interpretability of ML models for predicting multiple properties relevant to pharmaceutical and biological applications.
1 January 2026 in Digital Discovery
The central message of this Perspective is that sustainable ML requires not only designing more efficient algorithms but also transparent reporting, reusable datasets and models, informed benchmarking, and resource-aware community practices.
7 February 2025 in ChemRxiv
Biomolecular simulations have been paramount in advancing our understanding of the complex dynamics in biological systems. They have played a crucial role in various applications, including drug discovery and the molecular characterization of virus-host interactions. Despite their success, biomolecular simulations face inherent challenges due to the multiscale nature of biological processes, which involve intricate interactions across a wide range of length and time scales. All-atom molecular dynamics (AA-MD) provides detailed insights at atomistic resolution, yet it remains limited by computational constraints, capturing only short time scales and small conformational changes. In contrast, coarse-grained (CG) models extend simulations to biologically relevant time and length scales by reducing molecular complexity. However, CG models often sacrifice atomic-level accuracy, making the parameterization of reliable and transferable potentials a persistent challenge. This review discusses recent advancements in machine learning (ML)-driven biomolecular simulations, including the development of ML potentials with quantum-mechanical accuracy, ML-assisted backmapping strategies from CG to AA resolutions, and widely used CG potentials. By integrating ML and CG approaches, researchers can enhance simulation accuracy while extending time and length scales, overcoming key limitations in the study of biomolecular systems.
27 November 2025 in Advances In Space Research