Inferring Population Histories for Ancient Genomes Using Genome-Wide Genealogies.

Leo Speidel,Garrett Hellenthal,Pontus Skoglund,Robert W Davies,Lara Cassidy,Simon R Myers,Kelley Harris

doi:10.1093/molbev/msab174

Abstract

Ancient genomes anchor genealogies in directly observed historical genetic variation and contextualize ancestral lineages with archaeological insights into their geography and cultural associations. However, the majority of ancient genomes are of lower coverage and cannot be directly built into genealogies. Here, we present a fast and scalable method, Colate, the first approach for inferring ancestral relationships through time between low-coverage genomes without requiring phasing or imputation. Our approach leverages sharing patterns of mutations dated using a genealogy to infer coalescence rates. For deeply sequenced ancient genomes, we additionally introduce an extension of the Relate algorithm for joint inference of genealogies incorporating such genomes. Application to 278 present-day and 430 ancient DNA samples of >0.5x mean coverage allows us to identify dynamic population structure and directional gene flow between early farmer and European hunter-gatherer groups. We further show that the previously reported, but still unexplained, increase in the TCC/TTC mutation rate, which is strongest in West Eurasia today, was already present at similar strength and widespread in the Late Glacial Period ~10k−15k years ago, but is not observed in samples >30k years old. It is strongest in Neolithic farmers, and highly correlated with recent coalescence rates between other genomes and a 10,000-year-old Anatolian hunter-gatherer. This suggests gene-flow among ancient peoples postdating the last glacial maximum as widespread and localizes the driver of this mutational signal in both time and geography in that region. Our approach should be widely applicable in future for addressing other evolutionary questions, and in other species.

Highlights

Genetic variation is shaped through evolutionary processes acting on our genomes over hundreds of millennia, including past migrations, isolation by distance, mutation, or recombination rate changes, and natural selection
Relate and Colate Applied to 278 s Genome Diversity Project (SGDP) Moderns and 430 Ancients We inferred joint genealogies of 278 modern-day individuals of the SGDP and 14 previously published high coverage genomes of ancient individuals of >8x mean coverage, which we collectively rephase using Shapeit4 (Delaneau et al 2019) and the 1000 Genomes Project reference panel
The last decade has seen an explosion in the number of sequenced ancient genomes, uncovering remarkable stories of population replacements and admixture that are associated with dramatic shifts in lifestyle arounds the world (Skoglund and Mathieson 2018)

Summary

Introduction

Genetic variation is shaped through evolutionary processes acting on our genomes over hundreds of millennia, including past migrations, isolation by distance, mutation, or recombination rate changes, and natural selection. We introduce an extension to the Relate algorithm to enable the incorporation of samples of variable ages We use this approach to reconstruct joint genealogies of the Simon’s Genome Diversity Project (SGDP) data set (Mallick et al 2016) and 14 previously published high-coverage ancient humans covering diverse ancestries and sampled across the last 45k years (Fu et al 2014; Lazaridis et al 2014; GallegoLlorente et al 2015; Jones et al 2015; Broushaki et al 2016; Sikora et al 2017; de Barros Damgaard et al 2018; Gunther et al 2018; Sikora et al 2019; Cassidy et al 2020). A similar approach could be applied to other species

Methods

Results

Conclusion