Graph-based pangenome analysis uncovers structural and functional impacts of allopolyploidization events

The emergence of graph pangenomes has advanced the study of structural variants (SVs) across diverse species, yet the impact of complex ploidy on SV landscapes remains poorly explored. Brettanomyces bruxellensis, a yeast shaped by multiple allopolyploidization events, offers a compelling model due to its well-characterized genomic and phenotypic diversity. Here, we integrated subgenome-resolved assemblies with population-scale genomic and transcriptomic datasets to characterize SV distribution and functional impact at the species level. Leveraging a reference graph pangenome, we detected 5972 unique SVs across 1060 isolates. Allopolyploid genomes harbored significantly more SVs (320 on average) than diploids (116), reflecting hybridization's extensive impact on genome architecture. Functional association analysis revealed that 27.5% of SVs locally influence gene expression. Furthermore, a subset of SVs originating from ancestral divergence between subgenomes (hybridization-associated SVs) formed hotspots of structural diversity and disproportionately affect gene expression, with larger effect sizes than other SVs. These findings underscore allopolyploidization as a driver of structural and functional genomic diversification, highlighting the value of graph-based approaches in dissecting complex genome evolution.