Regio-Selectivity Prediction with a Machine-Learned Reaction Representation and On-the-Fly Quantum Mechanical Descriptors

Abstract

<div> <div> <div> <p>We introduce a new method that combines machine-learned reaction representation with selected quantum mechanical descriptors to predict regio-selectivity in general substitution reactions. We construct a reactivity descriptor database based on ab initio calculations of 130k organic molecules, and train a multi-task constrained model to calculate demanded descriptors on-the-fly. </p> </div> </div> </div>

An analysis of electrophilic aromatic substitution: a “complex approach”

“…Thus, with a view to advance the research in this field, it can be said that although a relatively small number of p-complexes have been tested to date, to our knowledge, the results presented here along with our conclusions can be considered to be encouraging, bearing in mind that EAS, as a transform, is still of great interest to many researchers with emerging new techniques. [101][102][103]…”

3D Computer Vision Models Predict DFT-Level HOMO-LUMO Gap Energies from Force-Field-Optimized Geometries

“…At the fundamental level, reactions proceed through 3D interactions, and we therefore expect that explicitly modeling the 3D shape of reaction components could lead to better performance for reactivity tasks. These include but are not limited to reaction yield [37,51], selectivity [49,15], and condition prediction [28], and even retrosynthesis planning [38].…”

Real-time prediction of 1H and 13C chemical shifts with DFT accuracy using a 3D graph neural network

“…These GNN-derived atomic descriptors impose low computational cost such that we anticipate future utility in related prediction tasks of organic reactivity and selectivity, for example in combination with other machine-learned representations. 87 Fig. 6 Screening and revising misassignment in NMRShiftDB.…”