arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2511.09571q-bio.QMcs.AI
收藏

基于通用智能的碎裂(GIF):一种峰标注谱图模拟框架

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

Department of Computer Science, Tufts University, Medford, MA 02155, USA(计算机科学系,塔夫茨大学,马萨诸塞州梅德福,02155,美国)

Margaret R. Martin, Soha Hassoun

AI总结:

针对代谢组学谱图标注率低的问题,提出基于通用智能的碎裂框架GIF,通过结构化提示引导LLM模拟质谱,实验证明其性能优于现有预训练模型与深度学习基线。

摘要:

尽管参考库不断扩充、计算工具持续迭代,代谢组学领域的进展仍受实测谱图标注率偏低的制约。近年来大语言模型(LLM)在各类生成与推理任务中表现优异,推动学界愈发关注其在质谱标注等特定领域科学难题中的应用。本文提出全新框架General Intelligence-based Fragmentation(GIF,基于通用智能的碎裂),通过结构化提示与推理引导预训练LLM完成谱图模拟。GIF采用标注机制、结构化输入输出、系统提示、指令式提示及迭代优化策略,为零散的临时提示方案提供了结构化替代方案,凸显了对LLM执行复杂科学任务时施加系统性引导的必要性。基于GIF,本文评估了当前通用LLM运用碎裂推理能力及微调后开展强度预测的表现,在衍生自MassSpecGym数据集的全新QA数据集MassSpecGym QA-sim上开展基准测试。实验显示,GPT-4o与GPT-4o-mini生成的模拟谱图与真实谱图的余弦相似度分别达0.36和0.35,表现优于GPT-5、Llama-3.1、ChemDFM等其他预训练模型——尽管GPT-5发布时间更新、ChemDFM为领域专用模型,同时GIF性能也超过多个深度学习基线模型。对GIF的评估表明,LLM不仅可用于谱图模拟,还能支撑人在回路工作流,实现分子碎裂过程的结构化可解释推理。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑