应用近红外光谱技术建立仿刺参3种组分的快速定量模型，收集475个仿刺参(Apostichopus japonicus)样品，以蛋白质、锌和硒3种成分指标为代表，采用便携式NIR光谱仪MicroNIR1700(JDSU, USA)扫描得到仿刺参个体干样3种成分的近红外光谱数据，在全光谱区间建立PLS定量分析模型。结果显示：蛋白质模型的预测决定系数(Rp2)、相对分析误差(RPD)分别为0.904、3.22，模型可用于大量样品快速检测应用。锌、硒模型的Rp2分别为0.498和0.526，RPD分别为1.41和1.96，表明锌、硒模型效果不理想，需要进一步优化。本研究初步建立了仿刺参干样样品蛋白质、锌、硒定量分析模型，进行了仿刺参干样多种品质相关指标同时快速测定方法的探索，可为仿刺参养殖品质性状筛选和刺参消费市场产品质量管控提供方法和依据。
A rapid quantitative model of 3 components of Apostichopus japonicus was established by using near infrared spectroscopy to collect 475 A. japonicus samples. With protein, Zn and Se as representative indicators, the portable NIR spectrometer MicroNIR1700 (JDSU, USA) was used for scanning to obtain the near infrared spectrum data of 3 components in the dry A. japonicus samples and establish PLS quantitative analysis model in full spectral range. The results showed that the protein model, with predictive decision coefficient (Rp2) of 0.904 and relative percent deviation (RPD) of 3.22, could be used to quickly detect a large number of samples. Zn and Se models, respectively with Rp2 of 0.498 and 0.526 and RPD of 1.41 and 1.96, were not ideal and should be further optimized. This study preliminarily established the quantitative analysis model of protein, Zn and Se in dry A. japonicus samples, explored the methods for quick determination of quality indexes of dry A. japonicus samples, and could provide method and basis for the screening of quality traits in breeding of A. japonicus and the quality control of products in consumer market.
