Statistical evaluation of alternative models of human evolution

Area of Science:

• Human Evolution
• Population Genetics
• Genomics

Background:

• Understanding recent human evolution requires distinguishing demographic effects from selection on genome diversity.
• Genetic data suggest a recent African origin for Homo sapiens, but debate persists regarding replacement versus interbreeding with archaic humans during expansion.

Purpose of the Study:

• To statistically assess the likelihood of different modern human evolution scenarios.
• To model recent human evolutionary history using extensive DNA data and simulations.

Main Methods:

• Analysis of DNA data from 50 nuclear loci in African, Asian, and Native American populations.
• Extensive simulations to compare probabilities of various evolutionary models (replacement, multiregional, assimilation).
• Bayesian analysis to estimate key demographic parameters under the best-supported model.

Main Results:

• The simple African replacement model with exponential growth showed the highest probability (78%) compared to alternative scenarios.
• Bayesian analysis under the replacement model estimated human origin around 141,000 years ago, out-of-Africa exit at 51,000 years ago, and Americas colonization at 10,500 years ago.
• This model explains both shallow ancestry in mitochondrial DNA (mtDNA) and Y-chromosomes, and deep lineages in autosomal loci, previously misinterpreted as archaic interbreeding.

Conclusions:

• The African replacement model provides the most robust explanation for recent human evolutionary history and genome diversity.
• The findings reconcile genetic data, supporting a complete replacement of archaic hominins without significant interbreeding.
• The model accurately predicts key divergence times and population colonization events, including the peopling of the Americas.