Abstract: To characterize the differences of flavor precursor substances in Coffea arabica beans from Pu’er, Yunnan, a total of 12 C. arabica germplasm resources from the Pu’er production area were systematically evaluated for genetic diversity and metabolic characteristics of flavor precursor by combining simple sequence repeats (SSR) molecular marker technology, analysis of key quality components, and integrated multi-omics strategies. The results showed that the genetic similarity coefficient of the tested materials was 0.72, and they could be divided into three clusters. Although the overall genetic background was relatively narrow, significant intra-species variation still existed. Quality analysis indicated that the contents of primary metabolites (proteins, lipids, carbohydrates, etc.) and secondary metabolites (caffeine, chlorogenic acid, etc.) differed significantly, and there were synergistic or antagonistic relationships among metabolites. Comparative proteomic and metabolomic analysis of two representative differential Catimor materials, FJ240001 and FJ240024, identified 459 differentially expressed proteins and 146 differentially accumulated metabolites, respectively, which were mainly enriched in primary metabolic pathways such as amino acid, lipid, and carbohydrate metabolism, and nine common pathways including ABC transporters and purine metabolism were also discovered. Random forest analysis selected 30 core biomarkers, and the correlation network revealed functionally independent metabolic modules formed by these markers. This study elucidates the metabolic foundation of flavor precursor formation in Pu’er C. arabica at the molecular level, providing a theoretical basis for targeted quality improvement and germplasm innovation.

Key words: Coffea arabica; flavor precursor substances; genetic diversity; omics; Pu’er