General Intelligence-based Fragmentation (GIF): A framework for peak-labeled spectra simulation

TL;DR

This paper introduces GIF, a structured framework leveraging large language models for simulating mass spectra in metabolomics, improving annotation accuracy and enabling explainable reasoning.

Contribution

GIF provides a systematic prompting approach for LLMs to simulate spectra, outperforming existing models and deep learning baselines in metabolomics applications.

Findings

GPT-4o achieves cosine similarity of 0.36 with true spectra

GIF outperforms several deep learning baselines

GIF enables human-in-the-loop workflows and explainable reasoning

Abstract

Despite growing reference libraries and advanced computational tools, progress in the field of metabolomics remains constrained by low rates of annotating measured spectra. The recent developments of large language models (LLMs) have led to strong performance across a wide range of generation and reasoning tasks, spurring increased interest in LLMs' application to domain-specific scientific challenges, such as mass spectra annotation. Here, we present a novel framework, General Intelligence-based Fragmentation (GIF), that guides pretrained LLMs through spectra simulation using structured prompting and reasoning. GIF utilizes tagging, structured inputs/outputs, system prompts, instruction-based prompts, and iterative refinement. Indeed, GIF offers a structured alternative to ad hoc prompting, underscoring the need for systematic guidance of LLMs on complex scientific tasks. Using GIF, we evaluate current generalist LLMs' ability to use reasoning towards fragmentation and to perform intensity prediction after fine-tuning. We benchmark performance on a novel QA dataset, the MassSpecGym QA-sim dataset, that we derive from the MassSpecGym dataset. Through these implementations of GIF, we find that GPT-4o and GPT-4o-mini achieve a cosine similarity of 0.36 and 0.35 between the simulated and true spectra, respectively, outperforming other pretrained models including GPT-5, Llama-3.1, and ChemDFM, despite GPT-5's recency and ChemDFM's domain specialization. GIF outperforms several deep learning baselines. Our evaluation of GIF highlights the value of using LLMs not only for spectra simulation but for enabling human-in-the-loop workflows and structured, explainable reasoning in molecular fragmentation.