Answers are grounded in this paper and cite their sources.
Flavonoid metabolism determines fruit coloration, nutritional quality, and stress adaptation in horticultural crops, making identification of biosynthetic components essential for synthetic biology and precision breeding. However, sequence-homology-based annotation methods remain prone to missed and incorrect annotations, particularly for modifying enzymes with low sequence similarity. To address this limitation, we developed DeepFlav ( https://cbi.gxu.edu.cn/KPGF/html/), a comprehensive resource and integrated computational platform for flavonoid biosynthesis gene analysis. DeepFlav integrates high-quality genome and proteome data from 1,031 plant species and annotated records for 630,951 genes encoding enzymes associated with flavonoid biosynthesis. Systematic analysis reveals a hierarchical evolutionary pattern in flavonoid pathways, with broadly conserved core enzymes and continuously expanding modifying enzymes. Differences among plant lineages are reflected in both the number of related genes and branch-specific pathway reconfiguration. Unlike conventional databases, DeepFlav does not rely solely on homology searches but implements an prediction framework based on the ESM-2 protein language model, AlphaFold-based three-dimensional structure modeling, and initial docking screening with AutoDock Vina. This design establishes an integrated annotation paradigm connecting sequence, structure, and function. The platform also reconstructs an interactive flavonoid metabolic network through literature mining. Using DeepFlav, we identified candidate genes with low sequence similarity to known flavonoid tailoring enzymes but high prediction confidence. In vitro validation of three candidate glycosyltransferases by LC-MS/MS confirmed their catalytic activity, demonstrating the reliability of DeepFlav for discovering low-homology flavonoid glycosyltransferases. DeepFlav provides a systematic resource and scalable framework for investigating plant flavonoid metabolism evolution, discovering functional modifying enzymes, and developing high-value natural products through synthetic biology.
No takes yet. Share an insight, caveat, or question.
Zhao et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: