Despite recent advances in population genomics, much remains to be elucidated with regard to East Asian population history. The Ainu, a hunter-gatherer population of northern Japan and Sakhalin island of Russia, are thought to be key to elucidating the prehistory of Japan and the peopling of East Asia. Here, we study the genetic relationship of the Ainu with other East Asian and Siberian populations outside the Japanese archipelago using genome-wide genotyping data. We find that the Ainu represent a deep branch of East Asian diversity more basal than all present-day East Asian farmers. However, we did not find a genetic connection between the Ainu and populations of the Tibetan plateau, rejecting their long-held hypothetical connection based on Y chromosome data. Unlike all other East Asian populations investigated, the Ainu have a closer genetic relationship with northeast Siberians than with central Siberians, suggesting ancient connections among populations around the Sea of Okhotsk. We also detect a recent genetic contribution of the Ainu to nearby populations, but no evidence for reciprocal recent gene flow is observed. Whole genome sequencing of contemporary and ancient Ainu individuals will be helpful to understand the details of the deep history of East Asians. KEYWORDS Jomon; Tibet; migration; demography T HE rapid development of genomic technologies has greatly enhanced our understanding of the history of modern human dispersal out of Africa and the peopling of different continents (Li et al. 2008;Green et al. 2010;Reich et al. 2010). However, the history of populations in East Asia, including Siberia, remains poorly understood even though they account for a large fraction of the human population (Stoneking and Delfin 2010;Oota and Stoneking 2011). There is still no clear consensus regarding basic questions such as when, where, and how many times modern humans migrated into East Asia and Siberia. For example, several previous studies based on contemporary samples concluded that one migration from south to north could explain the genetic structure of East Asians (Li et al. 2008; HUGO Pan-Asian SNP Consortium 2009). However, recent studies indicate that this scenario is too simplistic: a recent study detected western Eurasian ancestry in an individual in southern Siberia 24,000 years ago as well as a substantial contribution of this ancestry to the gene pool of Native Americans (Raghavan et al. 2014), and several studies detected a clear genetic differentiation between northeast Siberians and central-south Siberians, with the latter being more closely related to northeast Asians (Rasmussen et al. 2010;Fedorova et al. 2013). Indigenous high-altitude populations of the Tibetan plateau provide another example of East Asian populations that do not fit into the simple one migration hypothesis. Tibetans and Sherpa show a divergent history from lowland East Asian populations such as Han Chinese (Jeong et al. 2014), and their adaptive haplotype spanning the EPAS1 (endothelial PAS domain-containing prote...