Advancing plant DNA barcoding: integrating chloroplast genome sequencing, cryptic diversity discovery and machine learning

Accurate plant species identification underpins taxonomy, conservation, ecological monitoring, and the authentication of medicinal and food resources. While classical morphology-based approaches often struggle with cryptic or closely related taxa, DNA barcoding has emerged as a standardized molecular framework for species identification. In plants, core plastid markers such as rbcL and matK, together with nuclear regions like ITS and ITS2, have been widely adopted, yet species-level resolution remains limited in recently diverged or hybridizing lineages. Recent advances in high-throughput sequencing have enabled chloroplast genome sequencing and plastome-scale “super-barcoding,” substantially improving discriminatory power and facilitating the derivation of lineage-specific and mini-barcodes. Concurrently, multi-locus barcoding, metabarcoding, and environmental DNA (eDNA) approaches are revealing cryptic diversity and reshaping our understanding of plant community structure and species interactions. Emerging machine-learning methods further enhance barcode-based classification, reference-library curation, and integrative species delimitation. This review synthesizes developments in plastome-guided barcoding, cryptic diversity discovery, and data-driven analytics, outlining methodological advances, practical constraints, and future directions. We emphasize that continued expansion and rigorous curation of reference libraries, combined with transparent benchmarking of computational models, are essential for reliable, scalable, and genome-aware plant identification systems in the genomic era.