A genomic region associated with protection againstsevere COVID-19 is inherited from NeandertalsHugo Zeberga,b,1
and Svante Pääboa,c,1
aDepartment of Evolutionary Genetics, Max Planck Institute for Evolutionary Anthropology, D-04103 Leipzig, Germany; bDepartment of Neuroscience,Karolinska Institutet, SE-17177 Stockholm, Sweden; and cHuman Evolutionary Genomics Unit, Okinawa Institute of Science and Technology, Okinawa904-0495, Japan
Contributed by Svante Pääbo, January 22, 2021 (sent for review December 21, 2020; reviewed by Tobias L. Lenz and Lluis Quintana-Murci)
It was recently shown that the major genetic risk factor associatedwith becoming severely ill with COVID-19 when infected by severeacute respiratory syndrome coronavirus 2 (SARS-CoV-2) is inheritedfrom Neandertals. New, larger genetic association studies now allowadditional genetic risk factors to be discovered. Using data from theGenetics of Mortality in Critical Care (GenOMICC) consortium, weshow that a haplotype at a region on chromosome 12 associatedwithrequiring intensive carewhen infectedwith the virus is inherited fromNeandertals. This region encodes proteins that activate enzymes thatare important during infections with RNA viruses. In contrast to thepreviously described Neandertal haplotype that increases the risk forsevere COVID-19, this Neandertal haplotype is protective against se-vere disease. It also differs from the risk haplotype in that it has amore moderate effect and occurs at substantial frequencies in all re-gions of the world outside Africa. Among ancient human genomes inwestern Eurasia, the frequency of the protective Neandertal haplo-type may have increased between 20,000 and 10,000 y ago and againduring the past 1,000 y.
Neandertals | COVID-19 | OAS1 | SARS-CoV-2
Neandertals evolved in western Eurasia about half a millionyears ago and subsequently lived largely separated from the
ancestors of modern humans in Africa (1), although limited geneflow from Africa is likely to have occurred (2–5). Neandertals aswell as Denisovans, their Asian sister group, then became extinctabout 40,000 y ago (6). However, they continue to have a bio-logical impact on human physiology today through genetic con-tributions to modern human populations that occurred duringthe last tens of thousands of years of their existence (e.g., refs.7–10).Some of these contributions may reflect adaptations to envi-
ronments outside Africa where Neandertals lived over severalhundred thousands of years (11). During this time, they are likelyto have adapted to infectious diseases, which are known to bestrong selective factors that may, at least partly, have differedbetween sub-Saharan Africa and Eurasia (12). Indeed, severalgenetic variants contributed by archaic hominins to modern hu-mans have been shown to affect genes involved in immunity (e.g.,refs. 7, 8, 13, 14). In particular, variants at several loci containinggenes involved in innate immunity come from Neandertals andDenisovans (15), for example, toll-like receptor gene variantswhich decrease the susceptibility to Helicobacter pylori infectionsand the risk for allergies (16). Furthermore, proteins interactingwith RNA viruses have been shown to be encoded by DNA re-gions introgressed from Neandertals more often than expected(17), and RNA viruses might have driven many adaptive events inhumans (18).Recently, it was shown that a haplotype in a region on chromo-
some 3 is associated with becoming critically ill upon infection withthe novel severe acute respiratory coronavirus 2 (SARS-CoV-2)(19) and was contributed to modern humans by Neandertals (20).Each copy of this haplotype approximately doubles the risk of itscarriers requiring intensive care when infected by SARS-CoV-2. Itreaches carrier frequencies of up to ∼65% in South Asia and ∼16%
in Europe, whereas it is almost absent in East Asia. Thus, althoughthis haplotype is detrimental for its carriers during the currentpandemic, it may have been beneficial in earlier times in South Asia(21), perhaps by conferring protection against other pathogens,whereas it may have been eliminated in East Asia by negativeselection.A new study from the Genetic of Mortality in Critical Care
(GenOMICC) consortium, which includes 2,244 critically illCOVID-19 patients and controls (22), recently became available.In addition to the risk locus on chromosome 3, it identifies sevenloci with genome-wide significant effects located on chromo-somes 6, 12, 19, and 21. Here, we show that, at one of these loci,a haplotype associated with reduced risk of becoming severely illupon SARS-CoV-2 infection is derived from Neandertals.
Results and DiscussionA Neandertal Haplotype on Chromosome 12.We investigated whetherthe index single-nucleotide polymorphisms (SNPs), that is, theSNPs with the strongest association (Materials and Methods), atthe seven loci associated with risk of requiring intensive care uponSARS-CoV-2 infection on chromosomes 6, 12, 19, and 21 (22)harbor Neandertal-like alleles. To this end, we required that oneof the alleles of the index SNPs should match all three high-qualityNeandertals genomes, while being absent in the genomes of 108African Yoruba individuals [r2 > 0.80; the 1000 Genomes Project(23)]. None of the index SNPs for the loci on chromosomes 6, 19,and 21 fulfilled these criteria, whereas the locus on chromosome12 did.To further investigate this locus, we used data from the
COVID-19 Host Genetics Initiative [HGI; round 4 (24)]. We findthat the SNPs in the chromosome 12 locus associated withCOVID-19 hospitalization (P < 1.0e-5; Fig. 1) are in linkagedisequilibrium (LD) (r2 ≥ 0.8) in Europeans and form a haplotype
Significance
We show that a haplotype on chromosome 12, which is asso-ciated with a ∼22% reduction in relative risk of becoming se-verely ill with COVID-19 when infected by SARS-CoV-2, isinherited from Neandertals. This haplotype is present at sub-stantial frequencies in all regions of the world outside Africa.The genomic region where this haplotype occurs encodesproteins that are important during infections with RNA viruses.
Author contributions: H.Z. and S.P. designed research; H.Z. performed research; H.Z. an-alyzed data; and H.Z. and S.P. wrote the paper.
Reviewers: T.L.L. , Institut Pasteur; and L.Q.-M., Max Planck Institute forEvolutionary Biology.
The authors declare no competing interest.
This open access article is distributed under Creative Commons Attribution License 4.0(CC BY).1To whom correspondence may be addressed. Email: hugo.zeberg@ki.se or paabo@eva.mpg.de.
This article contains supporting information online at https://www.pnas.org/lookup/suppl/doi:10.1073/pnas.2026309118/-/DCSupplemental.
Published February 15, 2021.
PNAS 2021 Vol. 118 No. 9 e2026309118 https://doi.org/10.1073/pnas.2026309118 | 1 of 5
GEN
ETICS
Dow
nloa
ded
at U
nive
rsity
of T
exas
at A
rling
ton
on F
ebru
ary
22, 2
021
of ∼75 kb (chr12: 113,350,796 to 113,425,679; hg19). LD to theindex SNP of the GenOMICC study is given in SI Appendix, TableS1. Haplotypes of this length carrying alleles absent in Yoruba butpresent in Neandertals are likely to have been introduced intothe gene pool of modern humans due to interbreeding withNeandertals (25).To test whether the 75-kb haplotype is the result of gene flow
from Neandertals, we analyzed its relationship to present-dayand archaic genomes. To do this, we used the haplotypes seenmore than 10 times among the individuals in the 1000 GenomesProject (23) and the genome sequences of a ∼70,000-y-old Ne-andertal from Chagyrskaya Cave in southern Siberia (26), a∼50,000-y-old Neandertal from Vindija Cave in Croatia (27), a∼120,000-y-old Neandertal from Denisova Cave in southernSiberia (1), and a ∼80,000-y-old Denisovan individual from thesame site (28). Fig. 2 shows a phylogenetic tree estimating therelationships among these haplotypes. Among the 64 modernhuman haplotypes, eight form a monophyletic group with thethree Neandertal sequences.Genomic segments with similarity to Neandertal genomes may
either derive from common ancestors of the two groups that livedabout half a million years ago or be contributed by Neandertals tomodern humans by mixing between the two groups when they metless than 100,000 y ago (25). To test whether a segment of 75 kbmay have survived in this region of the genome since the commonancestor of the groups without being broken down by recombi-nation that affects chromosomes in each generation, we use apublished equation (29), a generation time of 29 y (30), a regionalrecombination rate of 0.80 cM/Mb (31), and a split time betweenNeandertals and modern humans of 550,000 y (1) followed byinterbreeding ∼50,000 y ago. Under these assumptions, in thisregion, segments of length 16.3 kb or longer are not expected toderive from the population ancestral to Neandertals and modern
humans (P = 0.05), making it highly unlikely that a 75-kb haplo-type does so (P = 8.2e-9). We thus conclude that the haplotypeentered the human gene pool from Neandertals. In agreementwith this, a previous study (32) has described gene flow fromNeandertals in this genomic region.
COVID-19 Protection and Geographic Distribution. We find that theindex variant of the protective haplotype in the GenOMICC study(rs10735079, P = 1.7e-8) matches all three Neandertal genomesavailable. The relative risk of needing intensive care is reduced by∼22% per copy of the Neandertal haplotype (under the rare diseaseassumption, odds ratio [OR] = 0.78, 95% CI 0.71 to 0.85). Asexpected given the phylogeny (Fig. 2), almost all of the allelescosegregating with the protective allele of the index SNP are foundin the Neandertal genomes (34 of 35 called SNPs; see SI Appendix,Table S2, which, in contrast to Fig. 1, includes data contributed by23andMe to HGI).Today, the haplotype is almost completely absent in African
populations south of the Sahara but exists at frequencies of ∼25 to30% in most populations in Eurasia (Fig. 3). In the Americas, itoccurs in lower frequencies in some populations of African ancestry,presumably due to gene flow from populations of European orNative American ancestry (33).
Putative Functional Variants. The Neandertal haplotype protectiveagainst severe COVID-19 on chromosome 12 contains parts or allof the three genes OAS1, OAS2, and OAS3, which encode oligoa-denylate synthetases. These enzymes are induced by interferons andactivated by double-stranded RNA. They produce short-chain pol-yadenylates, which, in turn, activate ribonuclease L, an enzyme thatdegrades intracellular double-stranded RNA and activates otherantiviral mechanisms in cells infected by viruses (reviewed byref. 34).To investigate which of these genes might be involved in pro-
tection against severe COVID-19, we plot the genomic location of
Fig. 1. Genetic variants associated with COVID-19 hospitalization at theOAS locus. Variants marked in red have P values less than 1e-5. In Europeans,they are in LD with the index variant (r2 ≥ 0.8), forming a haplotype (blackbar) with the genomic coordinates chr12: 113,350,796 to 113,425,679. Pvalues are from the HGI (24), excluding the 23andMe data for which onlysparse SNP data are available. The x axis gives hg19 coordinates; genes in theregion are indicated below. The three OAS genes are transcribed from left toright. Yellow dots indicate rs10735079 (right, the GenOMICC index SNP) andrs1156361 (left, typed by the Human Origins Array).
Fig. 2. Phylogeny relating DNA sequences associated with COVID-19 severityon chromosome 12. Haplotypes from three Neandertal genomes, the Deni-sovan genome, and haplotypes seen more than 20 times in individuals in the1000 Genomes Project are included. The colored area indicates haplotypes thatcarry the protective allele at rs1156361. The tree is rooted with the inferredancestral sequence from Ensembl (46). Six heterozygous positions in the ar-chaic genomes were excluded. Haplotypes XXIX and XXX are partially madeup of Neandertal-like DNA sequences due to recombination events.
2 of 5 | PNAS Zeberg and Pääbohttps://doi.org/10.1073/pnas.2026309118 A genomic region associated with protection against severe COVID-19 is inherited from
Neandertals
Dow
nloa
ded
at U
nive
rsity
of T
exas
at A
rling
ton
on F
ebru
ary
22, 2
021
the OAS genes below the P values for the SNPs associated withsevere COVID-19 (Fig. 1). While the association (P < 1.0e-5)overlaps all three OAS genes, the SNPs with the most significantassociations (P < 5.0e-8) are in OAS3. However, the high level ofLD and stochasticity in the associations make any conclusion re-garding causality based on P values tenuous.Nevertheless, there are alleles on the Neandertal haplotype
which stand out as potentially functionally important. One SNP(rs10774671) has been described as affecting a splice acceptorsite in OAS1 (35). The derived allele at this SNP, which is themost frequent allele in present-day humans, alters splicing ofOAS1 transcript such that several protein isoforms are producedinstead of the ancestral isoform which is preserved in Neander-tals (p46) (36). The latter, Neandertal-like isoform has higherenzymatic activity than the derived isoforms common in modernhumans (37). Outside Africa, the ancestral allele is present onlyin the context of the Neandertal haplotype, whereas, in Africa, itexists independently of this haplotype, presumably as a geneticvariant inherited from the common ancestors of modern humansand Neandertals that was lost in modern human populations thatleft Africa (35).
In addition to the splice acceptor site, the Neandertal haplotypecontains a missense variant (rs2660) in OAS1, a missense variant(rs1859330) and two synonymous variants (rs1859329 andrs2285932) in OAS3, and a missense variant in OAS2 (rs1293767).Three of these Neandertal-like variants are ancestral and occur inAfrica (rs2660, rs1859330, and rs1859329), whereas two are derivedin Neandertals (rs2285932 and rs1293767).Several SNPs on the chromosome 12 haplotype have previously
been studied with respect to their effects on other viral infections.The Neandertal-like splice acceptor variant has been associatedwith protection against West Nile Virus (rs10774671, OR = 0.63,95% CI 0.5–0.83) (38), and the Neandertal-like haplotype hasbeen associated with increased resistance to hepatitis C infections(39). Notably, the Neandertal missense variant in OAS1 (rs2660)(or variants in LD with this variant) has been shown to be asso-ciated with moderate to strong protection against SARS-CoV[OR = 0.42, 95% CI: 0.20 to 0.89 (40)], although this study waslimited in numbers of cases and controls. The SARS-CoV isclosely related to SARS-CoV-2, emerged in 2003, and caused amortality rate of ∼9% among infected individuals of all ages, andmuch higher rates of fatalities in older individuals (41). Finally, the
Fig. 3. Geographic distribution of the allele indicative of the Neandertal haplotype protective against severe COVID-19. Pie charts indicate minor allelefrequency in red at rs1156361. Frequency data are from the 1000 Genomes Project (23). Map source data are from OpenStreetMap.
Fig. 4. Frequencies across time of two Neandertal haplotypes associated with COVID-19 severity. Frequencies for rs1156361 at the OAS locus on chromosome12 (A) and rs10490770 at the chromosome 3 locus (B). Error bars indicate SE (Wilson scores). Time periods are indicated in years before present (bp). Ancientdata are from a compiled dataset (42), and present-day data are from the 1000 Genomes Project (23).
Zeberg and Pääbo PNAS | 3 of 5A genomic region associated with protection against severe COVID-19 is inherited fromNeandertals
https://doi.org/10.1073/pnas.2026309118
GEN
ETICS
Dow
nloa
ded
at U
nive
rsity
of T
exas
at A
rling
ton
on F
ebru
ary
22, 2
021
Neandertal versions of the OAS genes are expressed differently inresponse to different viral infections in cells in tissue culture interms of both expression levels and splice forms (35).
Haplotype Frequencies across Time. During the past few years,genome-wide data from thousands of prehistoric humans havebeen generated and compiled (42). This makes it possible to beginto directly gauge how frequencies of genetic variants have changedover time. Although this approach is still limited by the relativelysmall numbers of individuals and geographic regions for whichdata are available, we apply it here for the two Neandertal-derivedhaplotypes that affect the clinical outcomes upon infection withSARS-CoV-2.To tag the Neandertal OAS haplotype on chromosome 12, we
use an SNP (rs1156361) that carries a derived Neandertal-likeallele, is associated with the index variant of the GenOMICCstudy (r2 = 0.99 in Eurasia), and is typed by the Affymetrix HumanOrigins array used to study the majority of ancient human ge-nomes used here (42). Although this analysis is limited in that ittracks a single tag SNP, the fact that it is derived on the Nean-dertal lineage and in LD with the Neandertal haplotype makes thisanalysis feasible. We restrict the analysis to Eurasia and divide thedata into five time windows that vary between 20,000 and 2,000 yin length, to balance the number of genomes available while stillallowing potential differences in frequency to be discerned.Fig. 4A shows that the Neandertal OAS haplotype seems to have
occurred at frequencies below 10% prior to 20,000 y ago. Between20,000 and 10,000 y ago, the allele frequency was in the order of15%. Subsequently, it seems to have been present at frequencies ator slightly below 20% until 3,000 y to 1,000 y ago. Intriguingly, thecurrent allele frequency in Eurasia is ∼30%, suggesting that theNeandertalOAS haplotype may have increased in frequency relativelyrecently.To similarly estimate the frequency of the Neandertal risk
haplotype on chromosome 3 (20), we use the SNP rs10490770 thatfulfills the criteria applied above for the chromosome 12 haplo-type (Fig. 4B). Prior to 20,000 y ago, we find no carrier of the riskhaplotype among 16 genomes available. Among individuals wholived between 20,000 and 10,000 y ago and later, the haplotype ispresent in ∼10% until today, when it occurs at a frequency of∼12.5%. Thus, similar to the OAS locus, the Neandertal chro-mosome 3 locus, the frequency seems to be lower in the periodprior to 20,000 y ago than in the later periods. However, the dataare still scarce, making this observation preliminary. In contrast tothe OAS locus, there is no indication of any increase in the fre-quency of the Neandertal haplotype on chromosome 3 inhistorical times.We caution that the prehistoric data available are heavily bi-
ased toward western Eurasia and are still sparse, particularly forolder periods. However, additional data from ancient humanremains are rapidly being generated, making us confident that itwill soon be possible to identify loci that may have been thetargets of positive and negative selection, by studying allele fre-quencies over time in certain geographical regions while cor-recting for migration events that caused genome-wide shifts inallele frequencies.Despite theses caveats, it is interesting that the Neandertal-
derived OAS locus has recently increased in frequency in Eura-sia. This is compatible with previous work on the variationamong present-day populations (32, 35, 43) suggesting that thislocus has been positively selected. It is also compatible with
Denisovans having contributed a version of this locus, whichcarries ancestral variants, for example, at the slice acceptor site(rs10774671), to people in Oceania, where it occurs at substan-tial frequencies today (44).
Conclusions. A Neandertal haplotype on chromosome 12 is pro-tective for severe disease in the current SARS-CoV-2 pandemic.It is present in populations in Eurasia and the Americas at car-rier frequencies that often reach and exceed 50%. The ancestralNeandertal OAS locus variants may thus have been advanta-geous to modern humans throughout Eurasia, perhaps due toone or many epidemics involving RNA viruses, especially giventhat the Neandertal haplotype has been found to be protectivefor at least three RNA viruses (West Nile virus, hepatitis C virus,SARS-CoV). Supporting this notion, simulations have demon-strated that the Neandertal OAS haplotype has been underpositive selection in modern humans (35). Strikingly, the OAS1protein encoded by the modern human OAS haplotype is oflower enzymatic activity than the one encoded by the Neandertalhaplotype (37). This may have been advantageous at some pointin Africa, because loss-of-function mutations of the OAS1 locushave occurred numerous times among primates (45), suggestingthat the maintenance of OAS1 activity is costly to an organism.One may speculate that, when modern humans encountered newRNA viruses outside Africa, the higher enzymatic activity of theancestral variants that they acquired through genetic interactionswith Neandertals may have been advantageous.Intriguingly, there is evidence that the Neandertal-like OAS
haplotype may have recently increased in frequency in Eurasia(Fig. 4A), suggesting that selection may have positively affected theNeandertal-derived OAS locus in the last millennium. Futurestudies of human remains from historical times will clarify whether,and when, this occurred.
Materials and MethodsThe index variants for the seven novel loci (rs9380142, rs143334143, rs3131294,rs10735079, rs74956615, rs2109069, and rs2236757) were obtained fromGenOMICC (22). The regional summary statistics from the round 4 release ofthe metaanalysis carried out by the COVID-19 HGI (24) (https://covid19hg.org/results) was used to analyze the chromosome 12 locus (hospitalized vs. pop-ulation controls, i.e., “B2” phenotype, using all ancestries but not including the23andMe study, due to limited release of number of variants). LD was calcu-lated using LDlink 4.1, and alleles were compared to the archaic genomesusing tabix (HTSlib 1.10). The haplotype associated with protection againstsevere COVID-19 was investigated using phylogenetic software (PhyML 3.0),and the probability of observing a haplotype of a certain length or longer dueto incomplete lineage sorting was calculated as described (29). The present-day haplotypes were constructed by including all variable positions in the re-gion chr12: 113,350,796 to 113,425,679, excluding singletons. Haplotypes seenmore than 10 times were included in the phylogenetic analysis. The inferredancestral states at variable positions among present-day humans were takenfrom Ensembl. Genotypes of ancient genomes of modern humans wereobtained from a compiled database (42). Maps displaying allele frequencies ofdifferent populations were made using Mathematica 11.0 (Wolfram Research,Inc.) and OpenStreetMap data.
Data Availability.. Previously published data were used for this work (COVID-19HGI 1000 Genomes Project).
ACKNOWLEDGMENTS. We are indebted to the COVID-19 HGI for makingthe summary statistics of the genetic associations available and to the MaxPlanck Society and the NOMIS Foundation for funding.
1. K. Prüfer et al., The complete genome sequence of a Neanderthal from the Altai
Mountains. Nature 505, 43–49 (2014).2. M. Kuhlwilm et al., Ancient gene flow from early modern humans into Eastern Ne-
anderthals. Nature 530, 429–433 (2016).3. M. Meyer et al., Nuclear DNA sequences from the Middle Pleistocene Sima de los
Huesos hominins. Nature 531, 504–507 (2016).
4. C. Posth et al., Deeply divergent archaic mitochondrial genome provides lower time
boundary for African gene flow into Neanderthals. Nat. Commun. 8, 16046 (2017).5. M. Petr et al., The evolutionary history of Neanderthal and Denisovan Y chromo-
somes. Science 369, 1653–1656 (2020).6. T. Higham et al., The timing and spatiotemporal patterning of Neanderthal disap-
pearance. Nature 512, 306–309 (2014).
4 of 5 | PNAS Zeberg and Pääbohttps://doi.org/10.1073/pnas.2026309118 A genomic region associated with protection against severe COVID-19 is inherited from
Neandertals
Dow
nloa
ded
at U
nive
rsity
of T
exas
at A
rling
ton
on F
ebru
ary
22, 2
021
7. C. N. Simonti et al., The phenotypic legacy of admixture between modern humansand Neandertals. Science 351, 737–741 (2016).
8. M. Dannemann, J. Kelso, The contribution of Neanderthals to phenotypic variation inmodern humans. Am. J. Hum. Genet. 101, 578–589 (2017).
9. H. Zeberg et al., A Neanderthal sodium channel increases pain sensitivity in present-day humans. Curr. Biol. 30, 3465–3469.e4 (2020).
10. H. Zeberg, J. Kelso, S. Pääbo, The Neandertal progesterone receptor. Mol. Biol. Evol.37, 2655–2660 (2020).
11. F. Racimo, S. Sankararaman, R. Nielsen, E. Huerta-Sánchez, Evidence for archaicadaptive introgression in humans. Nat. Rev. Genet. 16, 359–371 (2015).
12. E. K. Karlsson, D. P. Kwiatkowski, P. C. Sabeti, Natural selection and infectious diseasein human populations. Nat. Rev. Genet. 15, 379–393 (2014).
13. L. Abi-Rached et al., The shaping of modern human immune systems by multiregionaladmixture with archaic humans. Science 334, 89–94 (2011).
14. H. Quach et al., Genetic adaptation and Neandertal admixture shaped the immunesystem of human populations. Cell 167, 643–656.e17 (2016).
15. M. Deschamps et al., Genomic signatures of selective pressures and introgression fromarchaic hominins at human innate immunity genes. Am. J. Hum. Genet. 98, 5–21(2016).
16. M. Dannemann, A. M. Andrés, J. Kelso, Introgression of Neandertal- and Denisovan-like haplotypes contributes to adaptive variation in human toll-like receptors. Am.J. Hum. Genet. 98, 22–33 (2016).
17. D. Enard, D. A. Petrov, Evidence that RNA viruses drove adaptive introgression be-tween Neanderthals and modern humans. Cell 175, 360–371.e13 (2018).
18. D. Enard, D. A. Petrov, Ancient RNA virus epidemics through the lens of recent ad-aptation in human genomes. Philos. Trans. R. Soc. Lond. B Biol. Sci. 375, 20190575(2020).
19. D. Ellinghaus et al., Genomewide association study of severe Covid-19 with respira-tory failure. N. Engl. J. Med. 383, 1522–1534 (2020).
20. H. Zeberg, S. Pääbo, The major genetic risk factor for severe COVID-19 is inheritedfrom Neanderthals. Nature 587, 610–612 (2020).
21. S. R. Browning, B. L. Browning, Y. Zhou, S. Tucci, J. M. Akey, Analysis of human se-quence data reveals two pulses of archaic Denisovan admixture. Cell 173, 53–61.e9(2018).
22. E. Pairo-Castineira et al.; GenOMICC Investigators; ISARICC Investigators; COVID-19Human Genetics Initiative; 23andMe Investigators; BRACOVID Investigators; Gen-COVID Investigators, Genetic mechanisms of critical illness in Covid-19. Nature,10.1038/s41586-020-03065-y (2020).
23. A. Auton et al.; 1000 Genomes Project Consortium, A global reference for humangenetic variation. Nature 526, 68–74 (2015).
24. COVID-19 Host Genetics Initiative, The COVID-19 Host Genetics Initiative, a globalinitiative to elucidate the role of host genetic factors in susceptibility and severity ofthe SARS-CoV-2 virus pandemic. Eur. J. Hum. Genet. 28, 715–718 (2020).
25. S. Sankararaman, N. Patterson, H. Li, S. Pääbo, D. Reich, The date of interbreedingbetween Neandertals and modern humans. PLoS Genet. 8, e1002947 (2012).
26. F. Mafessoni et al., A high-coverage Neandertal genome from Chagyrskaya Cave.Proc. Natl. Acad. Sci. U.S.A. 117, 15132–15136 (2020).
27. K. Prüfer et al., A high-coverage Neandertal genome from Vindija Cave in Croatia.Science 358, 655–658 (2017).
28. M. Meyer et al., A high-coverage genome sequence from an archaic Denisovan in-
dividual. Science 338, 222–226 (2012).29. E. Huerta-Sánchez et al., Altitude adaptation in Tibetans caused by introgression of
Denisovan-like DNA. Nature 512, 194–197 (2014).30. K. E. Langergraber et al., Generation times in wild chimpanzees and gorillas suggest
earlier divergence times in great ape and human evolution. Proc. Natl. Acad. Sci.
U.S.A. 109, 15716–15721 (2012).31. A. Kong et al., A high-resolution recombination map of the human genome. Nat.
Genet. 31, 241–247 (2002).32. F. L. Mendez, J. C. Watkins, M. F. Hammer, Neandertal origin of genetic variation at
the cluster of OAS immunity genes. Mol. Biol. Evol. 30, 798–801 (2013).33. A. R. Martin et al., Human demographic history impacts genetic risk prediction across
diverse populations. Am. J. Hum. Genet. 100, 635–649 (2017).34. U. Y. Choi, J.-S. Kang, Y. S. Hwang, Y.-J. Kim, Oligoadenylate synthase-like (OASL)
proteins: Dual functions and associations with diseases. Exp. Mol. Med. 47, e144(2015).
35. A. J. Sams et al., Adaptively introgressed Neandertal haplotype at the OAS locusfunctionally impacts innate immune responses in humans. Genome Biol. 17, 246
(2016).36. H. Li et al.; for UK Primary Sjögren’s Syndrome Registry, Identification of a Sjögren’s
syndrome susceptibility locus at OAS1 that influences isoform switching, protein ex-
pression, and responsiveness to type I interferons. PLoS Genet. 13, e1006820 (2017).37. V. Bonnevie-Nielsen et al., Variation in antiviral 2′,5′-oligoadenylate synthetase
(2‘5’AS) enzyme activity is controlled by a single-nucleotide polymorphism at a splice-
acceptor site in the OAS1 gene. Am. J. Hum. Genet. 76, 623–633 (2005).38. J. K. Lim et al., Genetic variation in OAS1 is a risk factor for initial infection with West
Nile virus in man. PLoS Pathog. 5, e1000321 (2009).39. M. K. El Awady et al., Single nucleotide polymorphism at exon 7 splice acceptor site of
OAS1 gene determines response of hepatitis C virus patients to interferon therapy.
J. Gastroenterol. Hepatol. 26, 843–850 (2011).40. J. He et al., Association of SARS susceptibility with single nucleic acid polymorphisms
of OAS1 and MxA genes: A case-control study. BMC Infect. Dis. 6, 106 (2006).41. M. D. Sørensen et al., Severe acute respiratory syndrome (SARS): Development of
diagnostics and antivirals. Ann. N. Y. Acad. Sci. 1067, 500–505 (2006).42. David Reich Lab, Allen Ancient DNA Resource (AADR): Downloadable genotypes of
present-day and ancient DNA data, version 42.4, https://reich.hms.harvard.edu/allen-
ancient-dna-resource-aadr-downloadable-genotypes-present-day-and-ancient-dna-
data. Accessed 19 April 2020.43. S. Yair, K. M. Lee, G. Coop, The timing of human adaptation from Neanderthal in-
trogression. bioRxiv, [Preprint] (2020). 2020.10.04.325183. Accessed 30 November2020.
44. F. L. Mendez, J. C. Watkins, M. F. Hammer, Global genetic variation at OAS1 provides
evidence of archaic admixture in Melanesian populations. Mol. Biol. Evol. 29,1513–1520 (2012).
45. C. M. Carey et al., Recurrent loss-of-function mutations reveal costs to OAS1 antiviral
activity in primates. Cell Host Microbe 25, 336–343.e4 (2019).46. A. D. Yates et al., Ensembl 2020. Nucleic Acids Res. 48, D682–D688 (2020).
Zeberg and Pääbo PNAS | 5 of 5A genomic region associated with protection against severe COVID-19 is inherited fromNeandertals
https://doi.org/10.1073/pnas.2026309118
GEN
ETICS
Dow
nloa
ded
at U
nive
rsity
of T
exas
at A
rling
ton
on F
ebru
ary
22, 2
021
