Machine learning classifies catalytic-reaction mechanisms

Danilo M. Lustosa is in the Department of Chemistry, Ben-Gurion University of the Negev, Be'er Sheva 84105, Israel. Anat Milo is in the Department of Chemistry, Ben-Gurion University of the Negev, Be'er Sheva 84105, Israel. The discovery of chemical reactions is influenced not only by how fast experimental data can be acquired, but also by how easily chemists can make sense of these data. Unravelling the mechanistic underpinnings of new catalytic reactions is a particularly intricate problem, often requiring expert knowledge of computational and physical organic chemistry. Nevertheless, it is important to study catalytic reactions because they represent the most efficient chemical processes.