The Rift Valley complex as a barrier to gene flow for Anopheles gambiae in Kenya

Recent studies of Anopheles gambiae, the principal mosquito vector of malaria in Africa, suggested that the eastern Rift Valley and its surrounding areas act as a barrier to gene flow. To quantify the unique effect of these areas on gene flow, we measured genetic variation within and between populations from each side of the Rift. Low differentiation was measured between populations on each side of the Rift (mean FST < 0.008, mean RST < 0.002). However, high differentiation was measured across the Rift (mean FST = 0.104; mean RST = 0.032). Genetic diversity within populations was lower in eastern populations, suggesting that the effective population sizes (Ne) of these populations were lower than those of western populations. We partitioned the overall differentiation across the Rift into three factors: variation in Ne between populations contributed 7-20%; distance contributed 10-30%, and the remainder, corresponding to the unique effect of the Rift was 50-80%. The Rift's effect was highly significant based on FST. The greater sensitivity of FST in measuring differentiation indicated that drift and not mutation generated the differences between populations. Restricted gene exchange across several hundred kilometers on the face of intense human transportation implies that active mosquito dispersal is the major form of migration, and that migration is a multistep process, where step length is relatively short.