Towards out-of-distribution generalizable predictions of chemical kinetics properties

TL;DR

This paper addresses the challenge of predicting chemical kinetic properties for unseen reactions and molecules by categorizing out-of-distribution scenarios, creating benchmarks, and evaluating current ML methods to identify gaps and opportunities.

Contribution

It introduces a framework categorizing OOD kinetic prediction into structure, condition, and mechanism levels, and provides comprehensive datasets and benchmarks for evaluating ML approaches in this context.

Findings

Current ML methods face significant challenges in OOD kinetic prediction.

Benchmark datasets reveal gaps in existing models' generalization capabilities.

Opportunities for developing more robust OOD prediction methods are identified.

Abstract

Machine Learning (ML) techniques have found applications in estimating chemical kinetic properties. With the accumulated drug molecules identified through "AI4drug discovery", the next imperative lies in AI-driven design for high-throughput chemical synthesis processes, with the estimation of properties of unseen reactions with unexplored molecules. To this end, the existing ML approaches for kinetics property prediction are required to be Out-Of-Distribution (OOD) generalizable. In this paper, we categorize the OOD kinetic property prediction into three levels (structure, condition, and mechanism), revealing unique aspects of such problems. Under this framework, we create comprehensive datasets to benchmark (1) the state-of-the-art ML approaches for reaction prediction in the OOD setting and (2) the state-of-the-art graph OOD methods in kinetics property prediction problems. Our results demonstrated the challenges and opportunities in OOD kinetics property prediction. Our datasets and benchmarks can further support research in this direction.