Fully understanding the chemicals in an herbal medicine remains a challenging task. Molecular networking (MN) allows to organize tandem mass spectrometry (MS/MS) data in complex samples by mass spectral similarity, which yet suffers from low coverage and accuracy of compound annotation due to the size limitation of available databases and differentiation obstacle of similar chemical scaffolds. In this work, an enhanced MN‐based strategy named diagnostic fragmentation‐assisted molecular networking coupled with in silico dereplication (DFMN‐ISD) was introduced to overcome these obstacles: the rule‐based fragmentation patterns provide insights into similar chemical scaffolds, the generated in silico candidates based on metabolic reactions expand the available natural product databases, and the in silico annotation method facilitates the further dereplication of candidates by computing their fragmentation trees. As a case, this approach was applied to globally profile the steroidal alkaloids in Fritillariae bulbus, a commonly used antitussive and expectorant herbal medicine. Consequently, a total of 325 steroidal alkaloids were discovered, including 106 cis ‐D/E‐cevanines, 142 trans ‐D/E‐cevanines, 29 jervines, 23 veratramines, and 25 verazines. And 10 of them were confirmed by available reference standards. Approximately 70% of the putative steroidal alkaloids have never been reported in previous publications, demonstrating the benefit of DFMN‐ISD approach for the comprehensive characterization of chemicals in a complex plant organism.

中文翻译：

---

诊断性碎片辅助质谱网络结合计算机去重复技术，可用于药用贝母中类固醇生物碱的深层注释。

全面了解草药中的化学物质仍然是一项艰巨的任务。分子网络（MN）允许通过质谱相似性来组织复杂样品中的串联质谱（MS / MS）数据，但是由于可用数据库的大小限制和相似化学物质的区分障碍，化合物标注的覆盖率和准确性较低脚手架。在这项工作中，引入了一种基于MN的增强策略，称为诊断碎片辅助分子网络结合计算机去重复（DFMN-ISD），以克服这些障碍：基于规则的碎片模式提供了对类似化学骨架的洞察力，基于代谢反应的硅候选物扩展了可用的天然产物数据库，电子注释方法通过计算候选者的碎片树来促进候选者的进一步去重复。作为一个案例，该方法被应用于对贝母（一种普遍使用的镇咳药和祛痰药）中的甾体生物碱进行整体分析。因此，总共发现了325种甾体生物碱，包括106种顺式-D / E-香兰素，142个反式-D / E-香兰素，29个牛er，23种藜芦胺和25种哒嗪。其中有10个已通过可用的参考标准确认。以前的出版物中从未报道过大约70％的假定甾体生物碱，这证明了DFMN-ISD方法对复杂植物有机体中的化学物质进行全面表征的好处。