The full title of the paper is “Full discovery of directional selection fulfills ancient DNA’s promise to explain human adaptation.” It really has great authors, including David Reich and Eric S. Lander, and many others from top schools. I read the paper, but I only partially understood it. In any case, here is the abstract:
We present a method for finding evidence of natural selection in ancient DNA time series data that exploits an untapped opportunity in previous scans: examining a consistent trend in allele frequency change over time. Applying this to 8433 West Eurasians who lived 14000 years ago and 6510 modern humans, we find an order of magnitude more significant genetic markers than previous studies: 347 independent loci with >99% probability of selection. Previous work has shown that classical strong sweeps driving beneficial adaptations have been rare over the broad span of human evolution, but over the past ten thousand years, hundreds of alleles have been affected by strong directional selection. Findings include an increase from ~0% to ~20% in 4000 years due to a higher risk of celiac disease HLA-DQB1; increase from ∼0% to 8% in 6000 years for blood type B; and variable selection in TYK2 Tuberculosis increased from ∼2% to 9% from ∼5500 to 3000 years ago before decreasing to 3%. We identify events of coordinated selection on alleles affecting the same trait, with a polygenic value today that predicts body fat percentages to decrease by about a standard deviation in ten thousand years, consistent with the “Thrifty Gene” theory that genes exist to store energy during meals. the shortage became dangerous after farming. We also identify a selection of combinations of alleles that today are associated with lighter skin color, lower risk of schizophrenia and mood swings, slower health decline, and increased measures related to cognitive functioning (scores on intelligence tests, household income, and years of schooling). These traits are measured in modern industrial societies, so which phenotypes were variable in the past is not clear. We estimate selection coefficients for 9.9 million variants, which allows to study how Darwin forces the couple in allelic effects and shapes the genes of complex traits.
I can report that nothing in their presentation seemed unreasonable or unfounded to me. But also the paper didn’t change my world view much? There are general Twitter speculations about how this might work for different groups, but note that the paper’s data aggregation methods actually require that different population groups (only Europe in the dataset) evolve in parallel and in similar ways over time. Without that thinking, all work falls apart.
Source link