Genomic connectivity and the spread of adaptive insecticide resistance alleles in Anopheles arabiensis from East Africa

Population connectivity and adaptive gene flow in disease vectors can shape the emergence and spread of insecticide resistance, with direct implications for control strategies such as insecticide spraying or the use of bed nets for malaria control. Using whole-genome sequencing, we first resolved the geographic population structure of the major but understudied malaria vector, Anopheles arabiensis, across the East African region, including the geographically diverse country of Ethiopia. We then assessed evidence for adaptive gene flow of insecticide resistance alleles across the region. Within Ethiopia, Central Rift Valley populations flanked by mountainous terrain were subject to restricted gene flow, although higher connectivity with the southwestern populations suggests an intermediate point of genetic exchange with the rest of Ethiopia. Anopheles arabiensis from western and northernmost Ethiopia were connected to populations from a similarly arid environment in Turkana - Kenya. Furthermore, broad-scale analysis revealed that populations from the rest of Kenya were connected with those from Uganda and Tanzania, but finer-scale analysis revealed more subtle structuring along the Rift Valley flanks, underscoring the role of landscape features in shaping patterns of gene flow. Adaptive gene flow analyses of diplotype clustering revealed that resistance alleles such as Cyp6aa/p and Gste2 copy number variants (CNV) were widely spread across East Africa despite the geographical population structuring we observed. At the Gste2 locus, An. arabiensis from Ethiopia and Kenya carried a newly annotated CNV spanning chromosome position 3R:28,596,832-28,606,222 linked to the non-synonymous SNP V47L, which was only at a low frequency in Kenya. Collectively, our findings demonstrate that An. arabiensis is subject to transboundary movement of resistance alleles, highlighting the need for coordinated cross-country vector management. Although areas of high connectivity may challenge genetic control technologies such as gene drives, more isolated populations may provide opportunities for targeted deployment. SignificanceMosquitoes movement across landscapes determines how insecticide resistance spreads, yet the genomic connectivity of Anopheles arabiensis populations in East Africa is understudied. Using whole genome data, we show that geographic features such as the Rift Valley restrict gene flow in Ethiopia, with western and northernmost Ethiopia connecting to Turkana, while Kenyan populations outside Turkana connect further south with Tanzania and Uganda. Despite these geographic barriers, resistance variants at genes such as Cyp6aa/p and Gste2 spread widely through adaptive gene flow, including a newly annotated CNV allele at Gste2 in Ethiopia and Kenya. These findings demonstrate that resistance alleles move across national borders, underscoring the need for coordinated regional vector management. They also highlight that while broader connectivity may limit genetic control tools such as gene drive, more isolated populations offer opportunities for targeted deployment.