How well can off-the-shelf LLMs elucidate molecular structures from mass spectra using chain-of-thought reasoning?

TL;DR

This paper evaluates the ability of large language models to interpret mass spectrometry data for molecular structure prediction using chain-of-thought reasoning, revealing their current limitations and potential for future improvements.

Contribution

It introduces a structured prompting framework and benchmark for assessing LLM reasoning on mass spectral data for molecular elucidation.

Findings

LLMs can generate syntactically valid structures

They often produce partially plausible but chemically inaccurate results

Current models lack reliable chemical reasoning capabilities

Abstract

Mass spectrometry (MS) is a powerful analytical technique for identifying small molecules, yet determining complete molecular structures directly from tandem mass spectra (MS/MS) remains a long-standing challenge due to complex fragmentation patterns and the vast diversity of chemical space. Recent progress in large language models (LLMs) has shown promise for reasoning-intensive scientific tasks, but their capability for chemical interpretation is still unclear. In this work, we introduce a Chain-of-Thought (CoT) prompting framework and benchmark that evaluate how LLMs reason about mass spectral data to predict molecular structures. We formalize expert chemists' reasoning steps-such as double bond equivalent (DBE) analysis, neutral loss identification, and fragment assembly-into structured prompts and assess multiple state-of-the-art LLMs (Claude-3.5-Sonnet, GPT-4o-mini, and Llama-3 series) in a zero-shot setting using the MassSpecGym dataset. Our evaluation across metrics of SMILES validity, formula consistency, and structural similarity reveals that while LLMs can produce syntactically valid and partially plausible structures, they fail to achieve chemical accuracy or link reasoning to correct molecular predictions. These findings highlight both the interpretive potential and the current limitations of LLM-based reasoning for molecular elucidation, providing a foundation for future work that combines domain knowledge and reinforcement learning to achieve chemically grounded AI reasoning.