目的 基于超分子“印迹模板”理论，整合匹配频数法、总量统计矩法与因子旋转法，结合体外抗肿瘤活性，解析鱼腥草Houttuynia cordata挥发油的药效物质基础，为其质量评价及潜在药效相关成分群筛选提供新思路。方法 收集58批不同产地鱼腥草挥发油（S1～S58），采用GC-MS法建立指纹图谱，以匹配频数法划分“印迹模板”为结构性“物质单元”；利用总量统计矩参数与信息熵值验证图谱整合前后特征的一致性；采用因子旋转法提取功能性“物质单元”并构建质量评价模型，对各批次样品评分；通过CCK-8法测定58批样品对人肺腺癌A549细胞的半数抑制浓度（half maximal inhibitory concentration，IC₅₀）值，采用皮尔逊相关性分析考察评分与IC₅₀值的关联。结果 58批样品经匹配频数法降维至34个结构性“物质单元”（A1～A34）；整合后平均色谱峰数及信息熵值显著降低（P＜0.01），而总量零阶矩、一阶矩、二阶矩及信息量无显著差异；因子旋转法提取8个公因子，综合评分结果显示，S16、S54等批次综合评分较高，S24、S27等批次综合评分较低；各批次IC₅₀值差异显著（73.6～269.9 nL/mL）；综合评分与IC₅₀值呈负相关（r＝-0.739，P＜0.01）；初步筛选出贡献度较大的A34、A31、A7等作为潜在药效成分群。结论 基于超分子“印迹模板”理论，利用匹配频数法划分结构性“物质单元”，总量统计矩法验证其科学性，结合因子旋转法构建综合评价模型，通过样品评分与抗肺癌活性的关联分析，初步筛选出鱼腥草挥发油中的关键结构性“物质单元”，提示了潜在的核心药效成分群，为后续精准筛选与验证提供了参考。

Objective Based on the supramolecular “imprinting template” theory, this study integrates the matching frequency method, total statistical moment, and factor rotation method, combined with in vitro anti-tumor activity, to elucidate the pharmacodynamic material basis of Yuxingcao (Houttuynia cordata) volatile oil, providing a new approach for its quality evaluation and screening of potential pharmacodynamic-related component groups. Methods Volatile oil samples from 58 batches of H. cordata collected from different origins (S1—S58) were analyzed by GC-MS to establish fingerprints. The matching frequency method was employed to categorize “imprinting template” into structural “material units”. Total-statistical-moment parameters and information entropy were applied to verify the consistency of fingerprint characteristics before and after integration. Functional “material units” were extracted via factor rotation to construct a quality evaluation model to score each batch. The half maximal inhibitory concentration (IC₅₀) values of the 58 volatile oil batches against human lung adenocarcinoma A549 cells were determined using the CCK-8 assay, and Pearson correlation analysis was conducted to examine the relationship between the comprehensive scores and IC₅₀ values. Results The 58 batches were effectively reduced to an “imprinting template” consisting of 34 structural “material units” (A1—A34) by the matching frequency method. After integration, the average number of chromatographic peaks and information entropy decreased significantly (P < 0.01), while no significant differences were observed in the total zero-order moment, first-order moment, second-order moment, or information content. Factor rotation extracted eight characteristic common factors. Comprehensive scoring results showed higher scores for batches such as S16 and S54, and lower scores for batches such as S24 and S27. Pharmacodynamic experiments revealed significant differences in IC₅₀ values among batches (73.6—269.9 nL/mL). Correlation analysis confirmed a negative correlation between comprehensive scores and IC₅₀ values (r = −0.739, P < 0.01). Accordingly, structural “material units” with high contributions, including A34, A31, and A7, were preliminarily screened as potential pharmacodynamic component groups. Conclusion Based on the supramolecular “imprinting template” theory, this study categorized structural “material units” using the matching frequency method and verified the scientific validity of this division with the total statistical moment. Subsequently, functional “material units” were extracted via factor rotation to construct a comprehensive evaluation model, and the sample scores were correlated with anti-lung cancer activity. This approach preliminarily screened key structural “material units” in H. cordata volatile oil, indicating potential core pharmacodynamic component groups, and providing a basis for subsequent precise screening and validation.

R283.6

国家自然科学基金项目（82274215）；湖南中医药大学2022年度校级“揭榜挂帅”专项（2022-XJJB001）；国家自然科学基金青年基金（82404815）；湖南省自然科学基金项目（2026JJ80968）；湖南中医药大学研究生科研创新项目（2025CX091）；湖南中医药大学本科生科研创新基金项目（2024BKS081）