Accurate, uncertainty-aware classification of molecular chemical motifs from multi-modal X-ray absorption spectroscopy

TL;DR

This paper introduces neural network ensemble classifiers that accurately identify chemical motifs from multi-modal X-ray absorption spectra, achieving high accuracy and reliable uncertainty quantification, with potential applications in molecular design.

Contribution

The work presents a novel multi-modal neural network ensemble approach for classifying chemical motifs with high accuracy and uncertainty estimation from XANES spectra.

Findings

Maximum average class-balanced accuracy of 0.99

Multi-modal data improves prediction performance

Method generalizes to molecular design applications

Abstract

Accurate classification of molecular chemical motifs from experimental measurement is an important problem in molecular physics, chemistry and biology. In this work, we present neural network ensemble classifiers for predicting the presence (or lack thereof) of 41 different chemical motifs on small molecules from simulated C, N and O K-edge X-ray absorption near-edge structure (XANES) spectra. Our classifiers not only reach a maximum average class-balanced accuracy of 0.99 but also accurately quantify uncertainty. We also show that including multiple XANES modalities improves predictions notably on average, demonstrating a "multi-modal advantage" over any single modality. In addition to structure refinement, our approach can be generalized for broad applications with molecular design pipelines.