Ancient DNA Study Reveals Accelerated Human Evolution in West Eurasia
A new study analyzing approximately 16,000 ancient and modern genomes from West Eurasia has identified 479 gene variants that show evidence of being shaped by natural selection over the past 10,000 years. The research, published in the journal Nature, suggests that the pace of human evolution in this region may have accelerated during this period, particularly after the advent of agriculture and during the Bronze Age.
The findings challenge a previous view that directional natural selection was rare in recent human history.
Study Scope and Method
The research, published on April 15, 2026, analyzed genomic data from West Eurasia, a region encompassing Europe and parts of western Asia. The dataset included 15,836 ancient human genomes, forming the largest such collection to date, alongside modern genomes for a total of approximately 16,000 samples.
Researchers developed a new statistical method called AGES (Ancient Genome Selection) to track evolutionary changes over an 18,000-year period. The method was designed to distinguish signals of natural selection from other evolutionary processes, such as random genetic drift, gene flow, and population migrations.
Traits Shaped by Selection
The analysis found that approximately 60% of the 479 identified variants correspond to known physical traits or disease susceptibilities in present-day populations.
Traits with increased frequency over time included gene variants associated with:
- Light skin pigmentation
- Red hair
- Resistance to HIV infection
- Resistance to leprosy (Hansen's disease)
- B blood type
Traits with decreased frequency over time included gene variants associated with:
- Male-pattern baldness
- Susceptibility to rheumatoid arthritis
Interpreting the Patterns
The study authors noted that while the DNA evidence indicates these variants were subject to selection, it does not definitively explain why. Researchers suggested the increase in light skin pigmentation frequency may reflect selection for increased vitamin D synthesis in regions with lower sunlight. They stated the reason for the rise in genes associated with red hair is less clear.
The research also found that selection pressures on some traits changed over time. For example:
- Genes linked to susceptibility to tuberculosis increased in frequency for several millennia, then decreased around 3,500 years ago.
- Genes linked to susceptibility to multiple sclerosis increased until about 2,000 years ago, then decreased.
Study first author Ali Akbari, a staff scientist at Harvard University, stated these patterns "likely reflect changes in environment or selective pressures over time; for example, the introduction of new pathogens."
Shifting Scientific Understanding
"Human evolution didn't slow down; we were just missing the signal," Akbari said. He explained that earlier methods, which relied on patterns in present-day genomes, led to the belief that directional selection was rare, but that new methods with large datasets allow detection of "small, consistent changes over time."
The study was co-led by David Reich, a population geneticist at Harvard Medical School. Reich stated the research shows "dramatic changes" and noted the Bronze Age was "an economically and culturally transformative time."
The article notes that some researchers have expressed skepticism regarding the scale of the findings, particularly results showing natural selection has affected gene variants underlying highly complex traits like mental illness and cognition.
Future Research Directions
The researchers have made their data and the AGES method freely available to other scientists. The research team plans to explore natural selection patterns in populations outside West Eurasia.
"What is likely to differ across regions is not whether selection occurred, but how local environments and cultural changes shaped it, including factors like diet, pathogens, and climate," Akbari said.
Source: The study "Ancient DNA reveals pervasive directional selection across West Eurasia" was authored by Akbari, A. et al. and published in Nature (2026), DOI: 10.1038/s41586-026-10358-1.