From MS/MS evidence to adaptive spectroscopic characterization: Reinforcement learning guided by candidate molecular structures

📄 arXiv: 2610.40264v1 📥 PDF

作者: Bowen Gao, Lei Zhu, Yiying Wang, Wenjie Yu

分类: cs.LG

发布日期: 2026-10-29

备注: 17 pages, 5 figures, 2 tables. Supporting information: 10 pages, 3 figures and 7 tables, provided as an ancillary PDF

💡 一句话要点

提出结构信息驱动的深度Q网络以优化小分子鉴定

🎯 匹配领域: 支柱二：RL算法与架构 (RL & Architecture) 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 小分子鉴定 深度Q网络 串联质谱 结构信息 强化学习 多模态光谱 化学特征

📋 核心要点

  1. 现有的小分子鉴定方法在串联质谱分析后，往往需要额外的测量来提高准确性，造成时间和资源的浪费。
  2. 本文提出的结构信息驱动深度Q网络（SI-DQN）通过利用现有的MS/MS证据，智能决策是否需要进一步表征及其方式，从而优化测量过程。
  3. 实验结果表明，SI-DQN在多个样本上显著提高了识别准确率，同时减少了额外测量次数，展现了其在小分子鉴定中的有效性。

📝 摘要（中文）

小分子从串联质谱（MS/MS）中识别可能需要补充测量。本文提出了一种结构信息驱动的深度Q网络（SI-DQN），利用现有的MS/MS证据来决定是否需要进一步表征、请求何种测量以及何时停止。该策略结合了光谱表示和候选分子的排名，以及候选分子化学特征和结构差异的得分加权描述符。在79,462个持出样本上进行评估，SI-DQN在正离子和负离子MS/MS起始的Top-1准确率分别提高至93.78%和93.31%。与基于规则的策略相比，SI-DQN在减少测量次数的同时显著提高了识别性能，提供了在MS/MS分析后利用候选结构间的模糊性指导进一步表征的定量基础。

🔬 方法详解

问题定义：本文旨在解决小分子鉴定中，串联质谱（MS/MS）分析后对进一步测量的需求决策问题。现有方法往往依赖于固定的测量流程，导致资源浪费和效率低下。

核心思路：论文提出的SI-DQN通过结合光谱表示和候选分子的化学特征，利用强化学习动态决策是否进行额外测量，优化了测量流程。

技术框架：SI-DQN的整体架构包括输入MS/MS证据、生成候选分子排名、计算得分加权描述符，并基于这些信息决定后续测量的必要性和类型。

关键创新：SI-DQN的创新在于将结构信息引入深度Q网络，使得模型能够在高结构不一致性情况下更有效地减少测量请求，显著提高了识别准确率。

关键设计：模型的损失函数结合了终端排名误差和测量成本，网络结构设计上采用了深度学习技术，以适应多模态光谱数据的处理需求。具体参数设置和网络层数在实验中经过调优，以实现最佳性能。

🖼️ 关键图片

fig_0

📊 实验亮点

实验结果显示，SI-DQN在正离子和负离子MS/MS起始的Top-1准确率分别提升至93.78%和93.31%，而额外测量次数仅为0.227和0.340，显著优于传统规则基线，复合损失降低幅度分别为19.3%和25.5%。

🎯 应用场景

该研究的潜在应用领域包括药物开发、环境监测和食品安全等领域，能够有效提高小分子鉴定的效率和准确性。通过减少不必要的测量，SI-DQN有助于节省时间和资源，推动相关领域的研究进展。

📄 摘要（原文）

Small-molecule identification from tandem mass spectrometry (MS/MS) may require complementary measurements. Herein, we propose a structure-informed deep Q-network (SI-DQN) that uses existing MS/MS evidence to decide whether further characterization is needed, which measurement to request, and when to stop. The policy combines spectral representations and candidate rankings with score-weighted descriptors of the candidates' chemical features and structural differences. The policy was evaluated on 79,462 held-out samples with simulated MS/MS, infrared (IR), and nuclear magnetic resonance (NMR) spectra from the Multimodal Spectroscopic Dataset. Evaluation used fixed hard candidate sets containing the correct structure and equal normalized acquisition costs. Averaged across three training runs, SI-DQN increased Top-1 accuracy from 80.10% to 93.78% for positive-ion MS/MS starts and from 70.44% to 93.31% for negative-ion starts. The corresponding mean additional measurement counts were only 0.227 and 0.340 per sample, respectively. Complete acquisition achieved 97.72% accuracy with four additions, whereas SI-DQN reduced the additional measurement count by 94.3% and 91.5%. Relative to a rule-based policy, SI-DQN reduced the composite loss combining terminal ranking error and acquisition cost by 19.3% and 25.5%. Adding structural information to DQN further improved accuracy while reducing requests across all three training runs. At comparable ranking uncertainty, SI-DQN increased request rates relative to DQN more strongly in groups with high structural disagreement, where larger reductions in composite loss were also observed. Together, these results support using the ambiguity among candidate structures after MS/MS analysis to guide further characterization, providing a quantitative basis for balancing identification performance against subsequent measurement effort.