
146 Regions Became 302: What AncestryDNA’s 2026 Update Actually Changed
2026年11月18日Tropical Ground Keeps Little Ancient DNA: A 5,000-Year Ceiling Measured from 300 Caribbean Skeletons
Ancient DNA research rests on an assumption that rarely makes it into a press release: the bone must still contain readable DNA. In Siberian permafrost or a European cave, that usually holds. In the hot, humid tropics it often does not.
On 9 September 2026, Royal Society Open Science published a systematic assessment that turns the question, how long does ancient DNA survive in the Caribbean, into usable numbers. The work was led by the Globe Institute at the University of Copenhagen, with Jazmín Ramos-Madrigal and Miren Iraeta-Orbegozo as first authors and Hannes Schroeder as corresponding author. Co-authors came from more than twenty institutions, among them the Max Planck Institute for Evolutionary Anthropology, the University of Havana, Leiden University and Arizona State University.
The sample size is the study’s strongest asset: 33 archaeological sites and more than 300 human skeletons from the Caribbean, spanning 3300 to 400 calibrated years before present, which covers most of the region’s human history. Every sample was shotgun-sequenced at low coverage to measure how much endogenous DNA could be recovered.
The first finding is a rate. Fitting decay curves to the sequencing data showed that DNA breaks down roughly four times faster in the Caribbean than in temperate regions and about fourteen times faster than in subpolar settings. On that basis, remains clearly older than about 5,000 years can be expected to yield nothing usable.
The second finding softens the first. Most of the Caribbean’s human past falls inside that 5,000-year window, so the ceiling removes the oldest chapter of the record without removing the record itself.
The real difficulty is variability. Preservation differed sharply between sites and within them. In a single cemetery, two skeletons a few metres apart might return more than 10% endogenous DNA in one case and almost nothing in the other. A good sample from a site therefore says nothing about whether the rest of that site will work.
The first practical lesson in the paper is to look before digging. Visual inspection of bone is imperfect, but it remains a useful first screen and reduces unnecessary destructive sampling.
The second is a ranking of material. The petrous part of the temporal bone and teeth, especially cementum, are the most reliable sources, while ribs and long-bone fragments should generally be avoided. This matches earlier experience that the petrous bone is the gold standard for ancient DNA, and it explains why field teams now set aside skulls and teeth first.
The third concerns setting. Caves and sheltered sites usually outperform open ground, and limestone bedrock with neutral to mildly alkaline sediments preserves DNA better than volcanic, iron-rich or acidic soils. Burial practice itself also leaves a mark.
The fourth is a rescue strategy. When endogenous DNA is very low, target capture can enrich it. The study compared 1240k capture across 82 Caribbean samples and found an average enrichment of about 280-fold, rising to roughly 606-fold in samples with less than 5% endogenous DNA. The authors caution, however, that enrichment reduces comparability between samples and cannot be mixed directly with shotgun data.
One more lesson is ethical. Ancient DNA analysis is destructive, so sampling plans should be discussed with local communities and stakeholders, and failed results should be published, otherwise later teams will spend the same samples for nothing.
For readers in East Asia, the value of this Caribbean study is methodological. Southern China and Southeast Asia share the same humid tropical climate, and the long-standing complaint that southern skeletons yield no DNA is exactly the accelerated decay measured here. The ranking of petrous bone and cementum first, caves over open ground, limestone over volcanic rock, transfers directly to sampling designs for southern cave sites. It is also why the 60,000-year-old bone fragments from Yunnan’s cave ultimately had no DNA to offer, and were identified as Denisovan through ancient proteomics instead: when heat and water take the DNA, some of the protein usually remains.
Excavation at the Golden Rock plantation site on the Caribbean island of St. Eustatius. How well bone survives in a hot, humid setting decides whether ancient DNA can be read at all. Source: AP Photo / Dick Drayer.





