Food Chemistry X· 2026Q1
Multi-omics and machine learning reveal the mechanisms underlying cultivar-driven flavor differentiation in fermented ciba chili
- 0citations
- Q1SCImago
- 2026year
Short summary
A multi-omics and machine learning approach identified 20 core metabolites and cultivar-specific bacterial communities that drive flavor differences in fermented ciba chili, with LCP favoring late-stage acid/ester enrichment, ICP showing earlier ester formation, and DCP accumulating alcohols, ketones, furans, and pyrazines.
AI-generated from the title and abstract; the full text is not read.
Key points
- Cultivar specificity profoundly shapes fermentation outcomes, leading to distinct volatile profiles (acids, esters, aldehydes, alcohols, ketones, furans, pyrazines) and metabolic enrichments (organic oxygen-containing compounds, organic acids, organoheterocyclic compounds, lipids, benzenoids).
- A random forest model identified 20 core discriminatory metabolites crucial for cultivar classification, linked to phenylpropanoid metabolism, lipid oxidation, purine cofactor metabolism, and nitrogen metabolism.
- Bacterial succession varied by cultivar, with DCP dominated by Lactiplantibacillus, LCP by Weissella, and ICP showing delayed Weissella dominance.
- Correlation analysis revealed cultivar-specific bacterial-metabolite networks influencing acidification, nitrogen metabolism, flavor precursors, complex aromas, and pungency.
AI-generated from the title and abstract; the full text is not read.
Abstract
Ciba chili quality is strongly influenced by raw material characteristics, but the cultivar-driven mechanisms underlying quality differentiation remain unclear. Ciba chili made from Inner Yellow New Generation (ICP), Lantern (LCP), and Devil's (DCP) peppers was comprehensively investigated using physicochemical analysis, GC-IMS, UHPLC-MS/MS untargeted metabolomics, 16S rRNA sequencing, and machine learning. Cultivar specificity profoundly shaped fermentation results. Volatile profiling revealed late-stage acid and ester enrichment in LCP, higher aldehyde retention alongside earlier ester formation in ICP, and an enrichment of alcohols, ketones, furans, and pyrazines in DCP. Furthermore, untargeted metabolomics showed that LCP, ICP, and DCP were enriched in organic oxygen-containing compounds, organic acids and organoheterocyclic compounds, and lipids and benzenoids, respectively. A random forest model utilizing these metabolites showed superior classification performance, and SHAP analysis further identified 20 core discriminatory metabolites contributing to cultivar classification. KEGG analysis traced flavor differentiation to phenylpropanoid metabolism, lipid oxidation, purine cofactor metabolism, and l -glutamine-mediated nitrogen metabolism. Cultivar-dependent bacterial succession was observed: DCP was dominated by Lactiplantibacillus , LCP by Weissella , and ICP showed delayed, less coordinated Weissella dominance. Crucially, correlation analysis revealed bacteria metabolite networks: LCP exhibited strong negative associations, ICP showed a balanced network linking acidification, nitrogen metabolism, and flavor precursors, while DCP displayed pronounced coupling associated with complex aromas and pungency. This study establishes an integrated multi-omics and machine learning framework to elucidate how chili cultivar drives quality differentiation, providing theoretical support for cultivar selection and flavor oriented industrial fermentation.
The authors' abstract, as published at the source. Food Chemistry X, 2026 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Food Science
Food ScienceAgricultural and Biological Sciences