Themed collection AI & ML in the Chemical Sciences

Integrating physics-based and deep learning methods advances protein–ligand modeling, boosting accuracy, scalability, and efficiency. This review surveys progress, integration strategies, challenges, and the outlook for AI-driven drug discovery.

AI-driven methods link spectral fingerprints to structures and properties, providing a foundation for the unified inverse design of functional substances and delivering interpretable insights into universal spectrum–structure–property relationships.

We offer ten diverse perspectives exploring the transformative potential of artificial intelligence (AI) in chemistry, highlighting many of the challenges we face, and offering potential strategies to address them.

This review offers a comprehensive overview of the development of machine learning potentials for molecules, reactions, and materials over the past two decades, evolving from traditional models to the state-of-the-art.

This article reviews computational tools for the prediction of the regio- and site-selectivity of organic reactions. It spans from quantum chemical procedures to deep learning models and showcases the application of the presented tools.

Enhanced sampling with ANI-2x enables rapid, systematic exploration of complex free energy landscapes to guide transition state refinement.

This work presents automated non-linear workflows for studying problems in low-data regimes alongside traditional linear models.

Nucleophilicity and electrophilicity are important properties for evaluating the reactivity and selectivity of chemical reactions.

We introduce HANNA, the first hybrid neural network model that strictly complies with all thermodynamic consistency criteria for predicting activity coefficients and outperforms current benchmark methods in terms of accuracy and applicability.

In this work we developed a highly accurate experimentally-based machine learning model with bespoke features that can predict the rate of ethene oligomerization propagation versus termination.

Leveraging multimodal large language models (MLLMs) to process multimodal data inputs and complex inter-modality data dependencies for automated (electro)chemical data mining from scientific literature.