Adaptive Challenges of Past and Future Invasion of <i>Drosophila suzukii</i> : Insights From Novel Genomic Resources and Statistical Methods Combining Individual and Pool Sequencing Data

Global change is accelerating biological invasions, making it crucial to understand how species adapt in new environments to improve management strategies. Genomic data provide valuable insights into adaptation through genotype-environment association (GEA) studies, which identify genes and biological processes tied to invasion success, and through geometric genomic offset (gGO) statistics, which estimate genetic (mal)adaptation to new environments. Here, we investigate genetic adaptation in the invasive pest Drosophila suzukii using novel genomic resources and statistical methods. We use a new chromosome-level genome assembly and data from 37 populations, combining publicly available and newly generated pooled and individual sequencing data, analysed with an enhanced version of BayPass software, tailored for such hybrid datasets. First, we identify genomic regions showing genetic differentiation between native and invasive populations. Then, using a GEA with 29 environmental covariates, we estimate the gGO between the source environments and the invaded areas, shedding light on the potential adaptive challenges D. suzukii faced during previous invasions. In addition, we estimate gGO for geographical areas not yet invaded to predict future invasion risks, and identify regions from which preadapted populations may originate. Our results reveal numerous genomic regions associated with the invasive status from genome scans. However, when considering broader patterns of adaptation to specific environmental variables through gGO analyses, we find that D. suzukii populations likely faced only limited adaptive challenges across their major invasion range, while certain uninvaded regions still remain at high risk of future invasion. Our study offers significant insights into D. suzukii adaptation and provides a practical population genomics framework to predict biological invasions, applicable to various species.