To utilize the nutritional composition of the meat and the mineral element composition of the shells of the Sinonovacula constricta to construct a reliable model for discriminating the origin and farming mode of the species.

The 4 major production areas in Zhejiang Province, namely Linhai, Sanmen, Ninghai(pond farming) and Xiangshan (mudflat farming), were selected as the research subjects. The quality indicators of the Sinonovacula constricta, including appearance traits, nutritional composition, amino acid composition, fatty acid composition and mineral element content, were systematically analyzed. Principal component analysis (PCA), partial least squares discriminant analysis (PLS-DA) and variable importance in the projection (VIP) were used to screen out the key characteristic parameters that affect the discrimination of the origin of the Sinonovacula constricta, and a discrimination model for different origins of the Sinonovacula constricta was constructed.

The quality analysis indicated that there were significant differences in the characteristics of the Sinonovacula constricta from different production areas and farming modes: The amino acid content (49.88 g/100 g) and the proportion of polyunsaturated fatty acids (41.46%) of the Sanmen (pond farming) Sinonovacula constricta were the highest; the content of trace elements in the Ninghai (pond farming) Sinonovacula constricta was relatively high; and the content of multiple nutritional indicators in the Xiangshan (mudflat farming) Sinonovacula constricta was relatively low.PCA, PLS-DA and VIP screened out 19 key characteristic parameters, including 5 kinds of minerals (iron, manganese, copper, calcium and zinc) in the soft tissue of the Sinonovacula constricta, 8 kinds of fatty acids(eicosapentaenoic acid, docosahexaenoic acid, etc.), and alanine; and 5 kinds of minerals (sodium, manganese, selenium, magnesium and calcium) in the shell. The constructed PLS-DA model had a good fit. The samples from pond farming and mudflat farming were clearly distinguished in the latent variable space; at the same time, the samples from the 3 adjacent pond farming production areas (Linhai, Sanmen and Ninghai) also showed independent clustering trends, indicating that the model could effectively capture the chemical fingerprint differences shaped by both the origin and the farming mode.

This study screens out 19 key characteristic parameters for the discrimination of the origin of the Sinonovacula constricta and constructed a discrimination model that can effectively distinguish the Sinonovacula constricta from different origins and farming modes (pond and mudflat)within Zhejiang Province. The model not only overcame the interference caused by the difference in farming modes but also accurately discriminate the Xiangshan (mudflat) production area, and shows good discrimination ability for the 3 adjacent pond farming production areas (Linhai, Sanmen and Ninghai). This provides a reliable technical means for the origin traceability of the Sinonovacula constricta under complex farming backgrounds.