Objective During the artificial breeding of Begonia, "false hybrids" frequently arise as a result of self-pollination or parthenogenesis, while traditional morphological identification and conventional molecular marker methods have limited capacity to distinguish them from true hybrids. To address these challenges, this study aims to develop a set of specific EPIC (Exon-primed intron-crossing) molecular barcodes suitable for the early and accurate identification of Begonia F₁ hybrids.

Method First, single-copy orthologous genes were identified using the whole-genome data of four representative Begonia species; highly variable regions were then screened, and intron-spanning amplification primers were designed. Subsequently, parents such as B. boweri cv. Tiger, together with six derived putative hybrid combinations, were selected as experimental materials for PCR amplification and Sanger sequencing. Finally, by aligning the sequences of parents and hybrid progenies, diagnostic single nucleotide polymorphisms (SNPs) and insertion/deletion (InDel) mutations were identified and quantified.

Result From 5 141 single-copy orthologous groups, four high-resolution EPIC barcode fragments (Bpe025368, Bpe012331.1, Bpe012357.1, and Bpe012390.1) were successfully screened and validated. Sequence alignment showed that true hybrids displayed characteristic biparental additive base peaks (overlapping peaks) at diagnostic sites, exhibiting typical biparental inheritance. For example, in the true hybrid B. heracleifolia cv. nigricans × B. cirrosa, the proportion of diagnostic SNP reached 100%. In contrast, false hybrids (such as the B. heracleifolia × B. heracleifolia cv. nigricans combination) exhibited homozygous single-parent inheritance, with an extremely low proportion of diagnostic SNP (0%~11.5%). Based on a comprehensive evaluation of amplification stability and sequencing clarity across all markers, Bpe012331.1 demonstrated the optimal identification efficacy and is recommended as the core marker.

Conclusion By exploiting the additive inheritance of nuclear genes, the EPIC barcode system developed in this study enables stable, rapid, and cost-effective identification of true and false Begonia F₁ hybrids. It provides a reliable molecular tool for germplasm resource conservation, new variety registration (DUS testing), and early-stage marker-assisted breeding.