@article {He2021.07.18.452867, author = {Guanglin He and Yunhe Zhang and Lan-Hai Wei and Mengge Wang and Xiaomin Yang and Jianxin Guo and Rong Hu and Chuan-Chao Wang and Xian-Qing Zhang}, title = {The genomic formation of Tanka people, an isolated {\textquotedblleft}Gypsies in water{\textquotedblright} in the coastal region of Southeast China}, elocation-id = {2021.07.18.452867}, year = {2021}, doi = {10.1101/2021.07.18.452867}, publisher = {Cold Spring Harbor Laboratory}, abstract = {Objectives Three different hypotheses proposed via the controversial evidence from cultural, anthropological and uniparental genetic analysis respectively stated that Tanka people probably originated from Han Chinese, ancient Baiyue tribe, or the admixture of them. Therefore, the genetic origin and admixture history of the Tanka people, an isolated {\textquotedblleft}Gypsies in water{\textquotedblright} in the coastal region of Southeast China, are needed to be genetically clarified.Materials and methods To elucidate the genetic origin of the Southeast Tanka people and explore their genetic relationship with surrounding indigenous Tai-Kadai and Austronesian people and Neolithic-to-historic ancients from the Yellow River Basin (YRB) and Fujian, we conducted a large-scale population genomic study among 1498 modern and ancient Eurasians, in which 73 Tanka and 4 Han people were first reported here. Both allele-shared and haplotype-based statistical methods were used here, including PCA, ADMIXTURE, f-statistics, ALDER, qpGraph-/TreeMix and qpAdm/qpWave, ChromoPainter and FineSTRUCTUREResults We found a specific genetic cline in PCA plots and detected the Tanka-specific homogeneous ancestry in model-based ADMIXTURE results, suggesting differentiated demographic history between Tanka and surrounding Hans. Formal tests based on sharing allele patterns showed a close relationship between Tanka people and Han Chinese, but the Tanka population harbored more southern East Asian ancestry related to Austronesian and Tai-Kadai people compared with southern Hans. Besides, the reconstructed differentiated demographic history revealed that southern Xinshizhou Tanka harbored more ancestry related to the Tai-Kadai or coastal ancient Neolithic to Bronze Age East Asians compared with northern Shacheng Tanka. The qpGraph-/TreeMix-based phylogenetic framework, qpAdm/qpWave-based admixture modeling and FineSTRUCTURE-based dendrogram among ancient northern and southern East Asians further demonstrated that the primary ancestry of modern Tanka derived from ancient millet farmers in the YRB with additional admixture from multiple southern East Asian sources.Discussion Sharing ancestry estimated from the f-statistics and sharing haplotypic landscape inferred from the ChromoPainter and FineSTRUCTURE showed that Southeast Tanka people not only had a close genetic relationship with both Northern Hans and YRB millet farmers but also possessed more southern East Asian ancestry related to Austronesian, Tai-Kadai and Hmong-Mien people. Our genomic data and fitted admixture models supported modern Tanka originated from ancient North China and obtained additional gene flow from ancient southern East Asians in the processes of southward migrations.Competing Interest StatementThe authors have declared no competing interest.}, URL = {https://www.biorxiv.org/content/early/2021/07/19/2021.07.18.452867}, eprint = {https://www.biorxiv.org/content/early/2021/07/19/2021.07.18.452867.full.pdf}, journal = {bioRxiv} }