Spectra to Structure: Deep Reinforcement Learning for Molecular Inverse Problem

Abstract

Spectroscopy is the study of how matter interacts with electromagnetic radiations of specific frequencies that has led to several monumental discoveries in science. The spectra of any particular molecule is highly information-rich, yet the inverse relation from the spectra to the molecular structure is still an unsolved problem. Nuclear Magnetic Resonance (NMR) spectroscopy is one such critical tool in the tool-set for scientists to characterise any chemical sample. In this work, a novel framework is proposed that attempts to solve this inverse problem by navigating the chemical space to find the correct structure that resulted in the target spectra. The proposed framework uses a combination of online Monte- Carlo-Tree-Search (MCTS) and a set of offline trained Graph Convolution Networks to build a molecule iteratively from scratch. Our method is able to predict the correct structure of the molecule ∼80% of the time in its top 3 guesses. We believe that the proposed framework is a significant step in solving the inverse design problem of NMR spectra to molecule.

Modern machine learning for tackling inverse problems in chemistry: molecular design to realization

“…The NMR spectra are incorporated as per-node information in the molecular graph, and the molecule is built iteratively by adding edges based on the probabilities returned by the neural network. Sridharan et al 139 used Monte Carlo tree search after framing the inverse problem as a Markov decision process. In this framework, value and prior models are pretrained using guided-MCTS runs incorporating substructure ** https://research.ibm.com/science/ibm-roborxn/ information.…”