Generative structural elucidation from mass spectra as an iterative optimization problem

TL;DR

This paper introduces FOAM, an iterative optimization workflow for structural elucidation from LC-MS/MS spectra, combining genetic algorithms and spectral simulation to improve metabolite annotation accuracy.

Contribution

The paper presents FOAM, a novel computational method that formulates structure elucidation as an iterative optimization problem, enhancing metabolite annotation from mass spectra.

Findings

FOAM performs well on NIST'20 and MassSpecGym datasets.

It can be used standalone or with existing inverse models.

Establishes iterative optimization as an effective paradigm.

Abstract

Liquid chromatography tandem mass spectrometry (LC-MS/MS) is a critical analytical technique for molecular identification across metabolomics, environmental chemistry, and chemical forensics. A variety of computational methods have emerged for structural annotation of spectral features of interest, but many of these features cannot be confidently annotated with reference structures or spectra. Here, we introduce FOAM (Formula-constrained Optimization for Annotating Metabolites), a computational workflow that poses structure elucidation from LC-MS/MS as an iterative optimization problem. FOAM couples a formula-constrained graph genetic algorithm with spectral simulation to explore candidate annotations given an experimental spectrum. We demonstrate FOAM's performance on the NIST'20 and MassSpecGym datasets as both a standalone elucidation pipeline and as a complement to existing inverse models. This work establishes iterative optimization as an effective and extensible paradigm for structural elucidation.