Ancient RNA virus epidemics through the lens of recent adaptation in human genomes

Abstract

Over the course of the last several million years of evolution, humans probably have been plagued by hundreds or perhaps thousands of epidemics. Little is known about such ancient epidemics and a deep evolutionary perspective on current pathogenic threats is lacking. The study of past epidemics has typically been limited in temporal scope to recorded history, and in physical scope to pathogens that left sufficient DNA behind, such as Yersinia pestis during the Great Plague. Host genomes, however, offer an indirect way to detect ancient epidemics beyond the current temporal and physical limits. Arms races with pathogens have shaped the genomes of the hosts by driving a large number of adaptations at many genes, and these signals can be used to detect and further characterize ancient epidemics. Here, we detect the genomic footprints left by ancient viral epidemics that took place in the past approximately 50 000 years in the 26 human populations represented in the 1000 Genomes Project. By using the enrichment in signals of adaptation at approximately 4500 host loci that interact with specific types of viruses, we provide evidence that RNA viruses have driven a particularly large number of adaptive events across diverse human populations. These results suggest that different types of viruses may have exerted different selective pressures during human evolution. Knowledge of these past selective pressures will provide a deeper evolutionary perspective on current pathogenic threats. This article is part of the theme issue 'Insights into health and disease from ancient biomolecules'.

Keywords: ancient epidemics; genomic adaptation; human evolution.

Figure 1.

Worldwide enrichment of iHS sweeps at VIPs compared to control non-VIPs. (a) Black line: observed fold enrichment at VIPs. Grey area: 95% confidence interval of the fold enrichment. Fold enrichments above 10 are represented at 10. When this happens, the confidence interval is not represented. However, the lowest edge of the confidence intervals not represented are all above 1. Red dots: bootstrap test p < 0.001 (see Methods). Dashed line: fold enrichment of 1, i.e. no enrichment. Fold enrichment (y-axis) is the number of VIPs in candidate sweeps divided by the average number of control non-VIPs in candidate sweeps. VIPs and non-VIPs in candidate sweeps are counted if they belong to the top x iHS genes (x-axis), where x is a rank threshold that slides from top 2000 to top 20, taking in total 25 values (2000; 1500; 1000; 900; 800; 700; 600; 500; 450; 400; 350; 300; 250; 200; 150; 100; 90; 80;70; 60; 50; 40;30; 25; 20). A fold enrichment of y = 3.51 at top x = 100 means that there are 3.51 times more VIPs in the top 100 iHS genes than control non-VIPs on average (over 1000 control sets of non-VIPs). There are in fact 80 VIPs in the iHS top 100, versus only 22.7 control non-VIPs. Eighty is high compared to 100 because of the summing over all 26 human populations. Specifically, a VIP or non-VIP counts as one in the top x if it is in the top x of at least one of the 26 populations. Note that counting the number of genes instead of counting the number of sweeps ignores the clustering of multiple genes in a single sweep, but that we account for this potential bias when estimating the whole rank threshold curve significance (see Methods). (b) Zoom-in on fold enrichment values from zero to two. (Online version in colour.)

Figure 2.

Sweep enrichment at RNA VIPs and DNA VIPs. Legend same as figure 1. (a) Sweep enrichment at RNA VIPs compared to other genes far from RNA VIPs (greater than 500 kb). The enrichment exceeds 10-fold at iHS top 30 and 25, and is represented at 10 without confidence intervals. (b) Sweep enrichment at DNA VIPs compared to other genes far from DNA VIPs (greater than 500 kb). Red dots: bootstrap test p < 0.001. Orange dots: p < 0.05. (Online version in colour.)

Figure 3.

Rationale of the test for confounding host intrinsic functions. If host functions cause the enrichment in adaptation at VIPs, then these over-represented host functions in VIPs concentrate the bulk of adaptation in VIPs but also in non-VIPs. This implies that non-VIPs in those host functions should exhibit more adaptation than non-VIPs in other host functions (right side of the table). If there is no difference (left side of the table), then host functions do not confound our analysis, and viruses are probably causal. (Online version in colour.)

Figure 4.

Power of iHS windows to detect various sweeps. (a) Incomplete selective sweeps from a de novo mutation that reached a 50% frequency. (b) Sweep from a standing, 5% standing variant at the start of selection and that reached a 50% frequency. (c,d) same as (a,b), respectively, but for sweeps that reached a 70% instead of 50% frequency. The y-axis represents the statistical power (true positive rate) at 0.1% false positive rate (FPR). The x-axis represents the range of simulated selection intensities, ranging from 2Ns = 50 to 2Ns = 1000, where N = 10 000 in our simulations and s is the selection coefficient. Blue curve: power with 50 kb windows. Orange curve: power with 1000 kb windows. Grey curve: ratio of the power with 1000 kb windows over the power with 50 kb windows. (Online version in colour.)