目的 建立海珠肉Pteria fucata的HPLC指纹图谱及多指标成分含量测定方法，结合化学计量学与电子鼻技术，系统评价不同加工方式与品质状态对其主要化学成分及气味特征的影响，并通过相关性分析揭示化学标志物与电子鼻响应之间的关联，从而多维度综合评价海珠肉的质量。方法 采用HPLC法建立海珠肉的化学指纹图谱，并定量测定谷氨酸、甘氨酸、牛磺酸及丙氨酸的含量。运用聚类分析（cluster analysis，CA）与主成分分析（principal component analysis，PCA）等化学计量学方法，综合评价不同样品间的化学差异。利用电子鼻采集样品的嗅觉信息，并通过Pearson相关性分析，构建化学成分与气味响应之间的关联模型，系统比较不同加工方式和品质状态样品的差异。结果 建立了海珠肉HPLC指纹图谱，共指认出4个共有峰，分别为牛磺酸、谷氨酸、甘氨酸与丙氨酸。其中，1、9、11、12号峰可用于区别鲜品微波（XW）和腐败品微波（FW）2种不同加工工艺；48个不同加工工艺及品质状态海珠肉样品的化学指纹图谱相似度在0.793～1.000，所建立的指纹图谱方法专属性良好，可区分不同品质海珠肉加工方式样品，用于海珠肉质量控制与评价。CA和PCA结果显示，不同品质加工工艺海珠肉样品有一定差异。含量测定表明，鲜品烘干（XH）工艺对多指标成分保留最优；电子鼻检测能有效区分不同品质及各加工方式样品，相关性分析揭示了谷氨酸与检测烷烃类及有机挥发物敏感的电子鼻传感器（通道）1、2、4呈显著正相关（P＜0.05），可作为与电子鼻响应信号关联的核心特征成分。结论 首次将HPLC指纹图谱、多指标含量测定与电子鼻技术联用，成功建立了一套稳定、可靠的综合品质评价方法体系。该方法体系具有良好的重复性与区分能力，能够从化学组成与挥发性气味2方面，实现对海珠肉品质特别是其新鲜与腐败状态的有效鉴别与综合评价。“鲜品烘干”工艺能最大程度保留谷氨酸等特征氨基酸成分，并维持良好的气味稳定性，是兼顾成分保留与品质控制的较优加工方式。

Objective To establish an HPLC fingerprint profile and multi-indicator component quantification method for Pteria fucata (PF), and systematically evaluate the effects of different processing methods by combining chemometrics and electronic nose technology, and quality states on its main chemical components and odor characteristics. Additionally, correlation analysis was conducted to reveal the relationships between chemical markers and electronic nose responses, enabling a multidimensional comprehensive evaluation of PF quality. Methods HPLC was used to establish the chemical fingerprint profile of PF, and the contents of glutamic acid, glycine, taurine, and alanine were quantitatively determined. Chemometric methods, including cluster analysis (CA) and principal component analysis (PCA), were applied to evaluate chemical differences among different samples. An electronic nose was employed to capture olfactory information, and Pearson correlation analysis was used to construct a model linking chemical components with odor responses. Differences among samples processed by various methods and quality states were systematically compared. Results An HPLC fingerprint profile of PF was established, identifying four common peaks corresponding to taurine, glutamic acid, glycine, and alanine. Peaks 1, 9, 11, and 12 could distinguish between fresh PF microwave (XW) and spoiled PF microwave (FW) processing methods. The similarity of chemical fingerprints among 48 samples with different processing methods and quality states ranged from 0.793 to 1.000. The established fingerprint method demonstrated good specificity and could differentiate samples of varying quality and processing methods, providing a reliable tool for quality control and evaluation. CA and PCA results indicated certain differences among PF samples of varying quality and processing methods. Quantitative analysis showed that the fresh PF drying (XH) process best preserved multiple indicator components. Electronic nose detection effectively distinguished samples of different quality and processing methods. Correlation analysis revealed that glutamic acid was significantly positively correlated (P < 0.05) with electronic nose channels 1, 2, and 4, which are sensitive to alkanes and organic volatile compounds, indicating its potential as a core feature component linked to electronic nose responses. Conclusion This study is the first to combine HPLC fingerprinting, multi-indicator component quantification, and electronic nose technology to establish a stable and reliable comprehensive quality evaluation system for PF. The method exhibits good repeatability and discriminative ability, enabling effective assessment of PF quality in terms of chemical composition and volatile odor characteristics, particularly distinguishing fresh from spoiled states. The fresh PF drying (XH) process maximizes the retention of key amino acids such as glutamic acid while maintaining odor stability, making it a superior processing method balancing component preservation and quality control.

R283.6

“尖峰”行动计划（广西重大专项计划）项目（桂科JF2503980037）；广西高峰学科中医学（中药学）（桂教科研〔2026〕2号）；广西中药药效研究重点实验室项目（桂科计字〔2005〕6号）；广西壮族自治区高校黄大年式教师团队“中药学传承创新教师团队”（桂教教师[2023]31号）