Abstract: Based on gas chromatography-mass spectrometry (GC-MS), metabolomics studies were conducted on 60 rice varieties growing in Jiansanjiang and Xiangshui, Heilongjiang. By applying the multivariate statistical analysis methods principal component analysis (PCA) and orthogonal partial least squares-discriminant analysis (OPLS-DA), we examined rice samples collected from the two growing areas. A total of 143 peaks were detected in the samples, and a total of 39 metabolites including amino acids, fatty acids, nucleotides, organic acids, polyols, sugars were identified. The effect of the geographical origin on rice metabolome was evaluated. The results showed significant differences in both the kinds and quantities of metabolites in rice samples from different growing areas. A total of 16 differential metabolites with significant changes (P < 0.05, variable importance in the projection (VIP) > 1) were identified. The results showed that GC-MS combined with multivariate statistical analysis could be used to study the differences in the metabolite profiles of rice samples collected from different growing regions. This study provides the basis for studying the geographical traceability of rice.

Key words: rice, gas chromatography-mass spectrometry (GC-MS), metabolomics, multivariate statistical analysis, differential metabolites