A series of ancient DNA studies from sites across China has identified at least one previously unknown East Asian genetic lineage that split off from other regional populations during or shortly after the last Ice Age. Researchers recovered genome-wide data from individuals buried near Beijing roughly 11,000 to 9,000 years ago, revealing a distinct northern branch that does not match any modern or previously sampled ancient group. The findings, drawn from multiple independent research teams working on sites from the Yellow River basin to Yunnan province and the Tibetan Plateau, are forcing a rethink of how human populations moved, mixed, and diverged across East Asia as glaciers retreated.
What is verified so far
The strongest single piece of evidence comes from the Donghulin site in western Beijing. A study published in Current Biology reports mitochondrial genomes from three individuals and genome-wide data from two individuals dated to approximately 11,000 to 9,000 years ago. The paper describes these people as carrying an “early-diverged northern East Asian lineage” that existed at or just after the close of the Pleistocene. That phrasing is significant: it means this group branched away from other East Asian populations early enough that its genetic signature looks fundamentally different from the ancestries researchers had catalogued before.
Independent evidence from southern China points in a similar direction. Genome-wide data from prehistoric Yunnan includes an Early Neolithic individual dated to roughly 7,100 years before present, described as carrying an unusually divergent “ghost” or basal lineage. The term “ghost” in population genetics refers to an ancestral group that has not been directly sampled but whose existence is inferred from statistical patterns in the DNA of its descendants. Finding a physical representative of such a lineage in Yunnan ties an abstract statistical signal to a real person buried in a specific place.
A separate line of research using mitochondrial genomes traces two major radiation events in East Asian maternal lineages: one during the Last Glacial Maximum, roughly 26,000 to 19,500 years ago, and another during the deglaciation period from about 19,000 to 11,500 years ago. That study links dispersals of matrilineal ancestry from northern coastal China as far as the Americas and Japan, establishing that post-glacial population movements were not local shuffles but long-range expansions with lasting genetic consequences.
Holocene-era genomes from northern China add further texture. Analysis of ancient individuals from the Yellow River, West Liao River, and Amur River basins, using standard ancient DNA panels such as 1240k and HumanOrigins with PCA, ADMIXTURE, and f-statistics workflows, shows that subsistence changes and migration were tightly linked across these regions. Farming transitions did not just change diets; they drove population replacements and admixture events that reshaped the genetic map of northern China.
A more recent study identifies a “previously unknown East Asian ancestry” described as a coastal lineage now associated with the Han Chinese, detected in ancient Shandong groups and introduced through admixture roughly 1,600 to 1,400 years before present. While that admixture event falls well after the Ice Age, the coastal ancestry itself may trace back much earlier, and its detection in Shandong fills a gap between the deep post-glacial lineages and the genetic profile of modern Han populations.
On the Tibetan Plateau, researchers report that populations over the past 5,100 years carry roughly 7 to 26 percent deeply diverged “ghost” ancestry, depending on the group and the statistical model applied, according to a study published in Science. That ghost component was not directly sampled in older datasets, which means it was invisible until analytical methods became sensitive enough to detect it. A related study links “Basal Asian Xingyi-related ancestry” from southwestern China to later Tibetan Plateau populations, with admixture influencing southern plateau groups after approximately 4,000 years before present.
What remains uncertain
The biggest open question is whether the Donghulin lineage from northern China and the Yunnan ghost ancestry from the south represent branches of the same deep split or entirely separate divergence events. No published study has yet produced direct admixture timelines linking these two signals. Researchers working on the Tibetan Plateau ghost ancestry rely on secondary modeling from citation trails rather than direct ancient DNA from the populations that would have carried the intermediate form of this ancestry. Until someone recovers and sequences ancient genomes from likely migration corridors, particularly the Yangtze River basin, the connection between northern and southern basal lineages remains a hypothesis rather than a demonstrated fact.
The coastal East Asian ancestry detected in Shandong groups raises its own set of questions. The admixture date of 1,600 to 1,400 years before present is well established, but the deeper origin of that coastal lineage is not. It could represent a surviving strand of post-glacial diversity that persisted along the coast while interior populations mixed and changed, or it could reflect a later migration from a source not yet identified. The data so far do not resolve this.
Coverage from Phys.org published in May 2025 provides institutional framing for some of these findings but does not add independent data beyond what the peer-reviewed papers contain. Readers should treat such coverage as a guide to the primary literature rather than as a standalone source of new claims.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
