Almost a decade ago, the sequencing of ancient DNA from archaic humans -Neanderthals and Denisovans -revealed that modern and archaic humans interbred at least twice during the Pleistocene. The field of human paleogenomics has now turned its attention towards understanding the nature of this genetic legacy in the gene pool of present-day humans. What exactly did modern humans obtain from interbreeding with Neanderthals and Denisovans? Were introgressed genetic material beneficial, neutral or maladaptive? Can differences in phenotypes among present-day human populations be explained by archaic human introgression? These questions are of prime importance for our understanding of recent human evolution, but will require careful computational modeling and extensive functional assays before they can be answered in full. Here, we review the recent literature characterizing introgressed DNA and the likely biological consequences for their modern human carriers. We focus particularly on archaic human haplotypes that were beneficial to modern humans as they expanded across the globe, and on ways to understand how populations harboring these haplotypes evolved over time.
Genome-wide patterns of archaic admixtureIn 2010, the first sequenced Neanderthal genome provided evidence for gene flow from Neanderthals into the ancestors of present-day non-Africans, around 50,000-60,000 years ago [1][2][3] . Since then, archaic human genomes have yielded ever more insightful discoveries. A few years later, a high coverage genome sequence from a Neanderthal found in the Altai mountains allowed researchers to pin down the proportion of Neanderthal ancestry in non-Africans to be ~2% [4] . In 2017, a second high coverage genome sequence from a Neanderthal in Croatia showed that this individual was more closely related to the introgressing Neanderthal population than the Altai Neanderthal, allowing researchers to detect even slightly higher levels of Neanderthal DNA [5] . In 2018, low-coverage genomes of five additional Neanderthals living between 39,000 and 47,000 years ago allowed a first glimpse at population structure in Neanderthals and showed indications of population turnover in late Neanderthal history [6] . But admixture between different human groups has not been limited to modern humans and Neanderthals. The genome sequence of a previously unknown group, the Denisovans (a sister group to Neanderthals), also contributed to the genomes of present-day people in Oceania, and, to a lower extent, to mainland East and South Asians [7][8][9][10][11][12] . Further admixture episodes have also been suggested, including gene flow from an unsampled "super-archaic" human group into Denisovans [4] , from eastern Neanderthals into Denisovans [4] and from modern humans into Neanderthals [13] (Figure 1). While the signals of shared ancestry between modern and archaic human groups are quite evident, the exact processes by which introgression occurred remain unclear. For example, higher levels of Neanderthal ancestry have been observed in...