Answers are grounded in this paper and cite their sources.
Perspective discusses challenges in predicting reaction properties using machine learning for synthesis planning, indicating future directions.
Machine learning is increasingly used to predict reaction properties such as barrier heights, reaction energies, rates, or yields, as well as the underlying molecular geometries, including transition state structures. While such predictions have the potential to provide mechanistic insight for high-impact applications such as synthesis planning and reaction optimization, the field remains at an early stage of development. This perspective discusses and critically assesses the current state of the art in (organic) reaction property prediction, highlighting the key limitations related to data availability and quality, molecular and transformation representations, and machine learning (ML) architectures used in both predictive and generative models. A special focus is given on current challenges and on possible paths forward toward efficient and accurate ML models for on-the-fly prediction of reaction energetics.
No takes yet. Share an insight, caveat, or question.
Landsheere et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: