A multiomics informed FT-NIR spectroscopy framework for nondestructive prediction of key flavor precursors in dried Rugiboletus extremiorientalis

Drying temperature influences flavor development of edible mushrooms, yet the metabolic basis of these changes and feasibility of rapid monitoring remain poorly understood. This study integrated HS-SPME-GC-MS, untargeted UPLC-Q-TOF-MS metabolomics, targeted LC-MS/MS and GC-MS quantification and FT-NIRS to characterize temperature-associated metabolic differences and identify key flavor precursors in Rugiboletus extremiorientalis (RE) during drying. Across five drying temperatures, 945 volatile compounds were tentatively assigned, with abundance differences in ketones, heterocyclic compounds, and terpenoids. Among 39 differential volatiles, 2(5H)-Furanone and 5-ethyl-3-hydroxy-4-methyl-2(5H)-furanone are the main sources of RE’s aroma profile. Untargeted metabolomics further characterized 2652 non-volatile metabolites, among which amino acids ranked prominently. Targeted analysis quantified 54 amino acids and 13 reducing sugars, and correlation analysis highlighted 11 compounds as key precursors linked to volatile profiles. WGCNA further revealed coordinated variation among these precursors under different drying conditions. FT-NIRS combined with PLSR provided reliable quantitative prediction for eight of the eleven key precursors with RPD above 2.0. L-leucine, L-glutamine, and L-threonine achieved R²p values of 0.993, 0.987, and 0.981, with RPD values of 2.241, 4.699, and 4.203. Overall, these findings advance understanding of temperature-associated metabolic differences in RE and demonstrate FT-NIRS potential for rapid, non-destructive screening of key flavor precursors during mushroom drying.