Evolution - Week 4: Human evolution

Monday, 14 October 13

Human evolution

Ancestors, relatives & major transitions

What about today?

Recent insights from genomics


Benton (2005) Fig 10.47


Relatives and recent ancestors

10 S C I E N T I F I C A M E R I C A N B E C O M I N G H U M A N

simple chewing surfaces—a feeding ap-paratus well suited to a diet of soft, ripe fruits. They also possessed shortened snouts, reflecting the reduced impor-tance of olfaction in favor of vision. His-tological studies of the teeth of Dryo-pithecus and Sivapithecus suggest that these creatures grew fairly slowly, as liv-ing great apes do, and that they probably had life histories similar to those of the great apes—maturing at a leisurely rate, living long lives, bearing one large off-spring at a time, and so forth. Other evidence hints that were they around to-day, these early great apes might have even matched wits with modern ones: fossil braincases of Dryopithecus indi-cate that it was as large-brained as a chimpanzee of comparable proportions. We lack direct clues to brain size in Siv-apithecus, but given that life history cor-relates strongly with brain size, it is like-ly that this ape was similarly brainy.

Examinations of the limb skeletons of these two apes have revealed addi-tional great ape–like characteristics. Most important, both Dryopithecus and Siva pith e cus display adaptations to

suspensory locomotion, especially in the elbow joint, which was fully extend-able and stable throughout the full range of motion. Among primates, this mor-phology is unique to apes, and it fi gures prominently in their ability to hang and swing below branches. It also gives hu-mans the ability to throw with great speed and accuracy. For its part, Dryo-pithecus exhibits numerous other adap-tations to suspension, both in the limb bones and in the hands and feet, which had powerful grasping capabilities. To-gether these features strongly suggest that Dryopithecus negotiated the forest canopy in much the way that living great apes do. Exactly how Sivapithecus got around is less clear. Some characteristics of this animal’s limbs are indicative of suspension, whereas others imply that it had more quadrupedal habits. In all likelihood, Sivapithecus employed a mode of locomotion for which no mod-ern analogue exists—the product of its own unique ecological circumstances.

The Sivapithecus lineage thrived in Asia, producing offshoots in Turkey, Paki stan, India, Nepal, China and South-

east Asia. Most phylogenetic analyses concur that it is from Si vapithecus that the living orangutan, Pon go pygmaeus, is descended. Today this ape, which dwells in the rain forests of Borneo and Sumatra, is the sole survivor of that suc-cessful group.

In the west the radiation of great apes was similarly grand. The recently discovered partial skeleton of Piero-lapithecus catalaunicus in northeastern Spain documents the earliest appear-ance of the lineage that includes the modern African apes, humans and our fossil relatives (australopithecines). Pier-olapithecus is closely related to Dryo-pithecus fontani, the ape found by Lar-tet, and may actually be the same animal. Over the next three million years or so, more specialized and modern-looking descendants would emerge. Within two million years four new species of Dryo-pithecus would evolve and span the re-gion from northwestern Spain to the Republic of Georgia. But where Dryo-pithecus belongs on the hominoid fam-ily tree has proved controversial. Some studies link Dryopithecus to Asian P

OR

TIA

SLO

AN

FAMILY TREE of hominoids encompasses the lesser apes (siamangs and gibbons), great apes (orangutans, gorillas and chimpanzees), and humans. Most

Miocene apes were evolutionary dead ends. But researchers have identifi ed a handful of them as candidate ancestors of living apes and humans. Proconsul, a primitive

Miocene ape, is thought to have been the last common ancestor of the living hominoids; Sivapithecus, an early great ape, is widely regarded as an orangutan forebear; and either

Dryopithecus or Ouranopithecus may have given rise to African apes and humans.

CATARRHINI

HYLOBATIDS

CERCOPITHECOIDS

PLATYRRHINI

SPIDER MONKEY MACAQUE SIAMANG GIBBON ORANGUTAN GORILLA HUMAN CHIMPANZEE

HOMINIDS

HOMINOIDS

SIVAPITHECUSPROCONSUL

OURANOPITHECUS

16 MYA19 MYA

DRYOPITHECUS

40 MILLION YEARS AGO

25 MYA

14 MYA

9 MYA6 MYA

Potential common ancestors(Miocene)

New world monkeys

Old world monkeys Lesser

apes

“Higher primates” = Old world monkeys + Apes

ApesGreat Apes

© Scientific American


Proconsul

Some ape-like featuresSome monkey-like features


East African Rift Valley











OR

TIA

SLO

AN





CATARRHINI

HYLOBATIDS

CERCOPITHECOIDS

PLATYRRHINI


HOMINIDS

HOMINOIDS


OURANOPITHECUS

16 MYA19 MYA

DRYOPITHECUS


25 MYA

14 MYA

9 MYA6 MYA



Major transitions in human evolution

• Bipedalism (down from the trees)

• Increased brain size

• Use of simple stone tools

• Fire, spears & other sophisticated tools (stone, bone...)

• Language, complex culture

• Agriculture...

In which order?


• Life in trees.

• Occassionally go down

• New context required going down more often?

Australopithecus africanus/!A. afarensis!

Ardipithecus ramidus!

Orrorin tugenensis!

Homo!

P. robustus!

Million!years!

Climate!

Cold!Warm!

Glacial cycles!

Arctic icecap!

Antarctic icecap!

WP!


Why bipedalism?

• Energy efficient locomotion (for distant food sources)

• Less exposure to sun?

• Free the hands? (for gathering/hunting?)

• Seeing farther : Finding food & avoiding predators?

• Sexual or anti-predator displays?

Habitat!fragmentation!

Climate!cooling!

Mid Miocene! Late Miocene!

Australopithecus africanus/!A. afarensis!

Ardipithecus ramidus!

Orrorin tugenensis!

Homo!

P. robustus!

Million!years!

Climate!

Cold!Warm!

Glacial cycles!

Arctic icecap!

Antarctic icecap!

WP!Monday, 14 October 13

Running

• sweating for thermoregulation.

• arched foot + achilles tendon

• head stabilization

• early Homo?

• first: improved scavenging.

• then persistence hunting












OR

TIA

SLO

AN





CATARRHINI

HYLOBATIDS

CERCOPITHECOIDS

PLATYRRHINI


HOMINIDS

HOMINOIDS


OURANOPITHECUS

16 MYA19 MYA

DRYOPITHECUS


25 MYA

14 MYA

9 MYA6 MYA



Most lineages went extinct

Proconsulidae

H. erectus

H. neanderthalensis

H. sapiens

Australopithecines



Australopithecines

Wikipedia


https://www.google.com/search?hl=en&safe=off&q=Australopithecines&spell=1&sa=X&ei=uf9NUcHAE8mxPLqPgOgE&ved=0CDEQvwUoAA


Taung child

Nature 19252.5 mya2.5 mya

Australopithecus afarensis 2.5 myaMonday, 14 October 13

http://en.wikipedia.org/wiki/Mya_(unit)




Lucy - Australopithecus afarensis

1978 3.2 myaMonday, 14 October 13

Australopithecines

0 3 000(km)

0 2 000(mi)Projection de Lambert azimutale équivalente

10° 20° 30° 40° 50° 60°0°10°20°30°

10° 20° 30° 40° 50° 60°0°10°20°30°

0°

10°

20°

30°

10°

20°

30°

0°

10°

20°

30°

10°

30°

A. GahriP. Boisei

A. Afarensis

A. Anamensis

A. Bahrelghazali

A. Africanus

P. Aethiopicus

P. Robustus (Crassidens)

Wikipedia

Brain size: 35% of modern human





Evidence for bipedalism in Australopithecines




• Pelvis short & broad (like humans), not long & narrow (like gorilla)• Hip & walking muscles arranged like in a bipedal organism• Femur angled as in humans, not straight as in chimps• Feet



Fossilized tracks at Laetoli (Tanzania)

Footprints preserved in volcanic ash from: 3 hominids (Australopithecus afarensis)Numerous other mammals

3.6Mya


Tool use?

• generally: only simple tools (similarly to current non-human great apes).

• but Australopithecus garhi (2.5 mya) may have made stone tools.


Summary: Australopithecines

• Major group of early bipedal hominids (4mya to 1 mya)

• Small brains

• Only in Africa

• Many forms/species




Proconsulidae

Australopithecines

H. erectus

H. neanderthalensis

H. sapiens


Homo


Homo habilis


Tool use

Chimps and other animals may use objects as tools.

H. habilis!H. sapiens! Australopithecine!

H. habilis made tools

ScrapingCutting


Stages of human evolution are defined by the style and sophistication of stone tools….

e.g.: •Oldowan (2.5-1.5 mya)•Achuelian (1.5-0.2 mya)


Oldowan toolsHammerstone Choppers

Scraper Flakes


Brain sizes increase



Out of Africa - H. erectus


Acheulian tools

Handaxe

Handaxes! Cleaver!

Pick!

Trimming flakes!Scraper!


Nariokotome/Turkana boy

Found 1984 in Kenya. From1.5mya

H. erectus


H. erectus lifestyle

• Stone tools (Acheulian)

• Fire

• Sociality

• Hunting • …language?



Homo floresiensis “The Hobbit”

H. florensis vs. H. sapiens skull


Nature (2004) vol. 431, 1043-1044Monday, 14 October 13










OR

TIA

SLO

AN





CATARRHINI

HYLOBATIDS

CERCOPITHECOIDS

PLATYRRHINI


HOMINIDS

HOMINOIDS


OURANOPITHECUS

16 MYA19 MYA

DRYOPITHECUS


25 MYA

14 MYA

9 MYA6 MYA


New world monkeys

Old world monkeys Lesser

apes

“Higher primates” = Old world monkeys + Apes

ApesGreat Apes



Major transitions in human evolution

• Bipedalism (down from the trees)

• Increased brain size

• Use of simple stone tools

• Fire, spears & other sophisticated tools (stone, bone...)

• Language, complex culture

• Agriculture...



Proconsulidae

Australopithecines

H. erectus

H. neanderthalensis

H. sapiens


•


Neanderthal

600,000-30,000 years ago Monday, 14 October 13


Burial ritual?


Neanderthals - Summary

• Neanderthals were morphologically and genetically distinct from early H. sapiens

• disappeared after H. sapiens arrived - possibly because they were culturally less advanced.



Proconsulidae

Australopithecines

H. erectus

H. neanderthalensis

H. sapiens


H. sapiens out of Africa

• 50,000 years ago: fully “modern” with language, music, advanced social intelligence, strategic planning etc.

• 70,000 years ago: began migrating out of Africa• Simultaneous decline of other Homo species (erectus,

neanderthalensis...): competition?• Superior cooperation & learning due to language?

• Agriculture ~ 10,000 years ago



Burial ritual in early H. sapiens

• At Sungir, Russia, around 28,000 years ago

• A 60 year old buried with an elaborate collection of beads, necklaces and bracelets


Examples of early H. sapiens tools


Lascaux - 35000 years ago

Lion man, Ulm - 40,000 years ago

Flute - 36,000 years ago



Recent insights from genomics


A Draft Sequence of theNeandertal GenomeRichard E. Green,1*†‡ Johannes Krause,1†§ Adrian W. Briggs,1†§ Tomislav Maricic,1†§Udo Stenzel,1†§ Martin Kircher,1†§ Nick Patterson,2†§ Heng Li,2† Weiwei Zhai,3†||Markus Hsi-Yang Fritz,4† Nancy F. Hansen,5† Eric Y. Durand,3† Anna-Sapfo Malaspinas,3†Jeffrey D. Jensen,6† Tomas Marques-Bonet,7,13† Can Alkan,7† Kay Prüfer,1† Matthias Meyer,1†Hernán A. Burbano,1† Jeffrey M. Good,1,8† Rigo Schultz,1 Ayinuer Aximu-Petri,1 Anne Butthof,1Barbara Höber,1 Barbara Höffner,1 Madlen Siegemund,1 Antje Weihmann,1 Chad Nusbaum,2Eric S. Lander,2 Carsten Russ,2 Nathaniel Novod,2 Jason Affourtit,9 Michael Egholm,9Christine Verna,21 Pavao Rudan,10 Dejana Brajkovic,11 !eljko Kucan,10 Ivan Gu"ic,10Vladimir B. Doronichev,12 Liubov V. Golovanova,12 Carles Lalueza-Fox,13 Marco de la Rasilla,14Javier Fortea,14¶ Antonio Rosas,15 Ralf W. Schmitz,16,17 Philip L. F. Johnson,18† Evan E. Eichler,7†Daniel Falush,19† Ewan Birney,4† James C. Mullikin,5† Montgomery Slatkin,3† Rasmus Nielsen,3†Janet Kelso,1† Michael Lachmann,1† David Reich,2,20*† Svante Pääbo1*†Neandertals, the closest evolutionary relatives of present-day humans, lived in large parts of Europeand western Asia before disappearing 30,000 years ago. We present a draft sequence of the Neandertalgenome composed of more than 4 billion nucleotides from three individuals. Comparisons of theNeandertal genome to the genomes of five present-day humans from different parts of the worldidentify a number of genomic regions that may have been affected by positive selection in ancestralmodern humans, including genes involved in metabolism and in cognitive and skeletal development.We show that Neandertals shared more genetic variants with present-day humans in Eurasia than withpresent-day humans in sub-Saharan Africa, suggesting that gene flow from Neandertals into theancestors of non-Africans occurred before the divergence of Eurasian groups from each other.

The morphological features typical of Nean-dertals first appear in the European fossilrecord about 400,000 years ago (1–3).

Progressively more distinctive Neandertal formssubsequently evolved until Neandertals disap-peared from the fossil record about 30,000 yearsago (4). During the later part of their history,Neandertals lived in Europe and Western Asiaas far east as Southern Siberia (5) and as farsouth as the Middle East. During that time, Nean-dertals presumably came into contact with ana-tomicallymodern humans in theMiddle East fromat least 80,000 years ago (6, 7) and subsequentlyin Europe and Asia.

Neandertals are the sister group of all present-day humans. Thus, comparisons of the humangenome to the genomes of Neandertals andapes allow features that set fully anatomicallymodern humans apart from other hominin formsto be identified. In particular, a Neandertal ge-nome sequence provides a catalog of changesthat have become fixed or have risen to highfrequency in modern humans during the lastfew hundred thousand years and should beinformative for identifying genes affected bypositive selection since humans diverged fromNeandertals.

Substantial controversy surrounds the questionof whether Neandertals interbred with anatomi-cally modern humans. Morphological featuresof present-day humans and early anatomicallymodern human fossils have been interpreted asevidence both for (8, 9) and against (10, 11) ge-netic exchange between Neandertals and the pre-

sumed ancestors of present-day Europeans.Similarly, analysis of DNA sequence data frompresent-day humans has been interpreted as evi-dence both for (12, 13) and against (14) a geneticcontribution by Neandertals to present-day hu-mans. The only part of the genome that has beenexamined from multiple Neandertals, the mito-chondrial DNA (mtDNA) genome, consistentlyfalls outside the variation found in present-dayhumans and thus provides no evidence for inter-breeding (15–19). However, this observationdoes not preclude some amount of interbreeding(14, 19) or the possibility that Neandertals con-tributed other parts of their genomes to present-day humans (16). In contrast, the nuclear genomeis composed of tens of thousands of recombin-ing, and hence independently evolving, DNA seg-ments that provide an opportunity to obtain aclearer picture of the relationship between Nean-dertals and present-day humans.

A challenge in detecting signals of gene flowbetween Neandertals and modern human ances-tors is that the two groups share common ances-tors within the last 500,000 years, which is nodeeper than the nuclear DNA sequence variationwithin present-day humans. Thus, even if no geneflow occurred, in many segments of the genome,Neandertals are expected to be more closely re-lated to some present-day humans than they are toeach other (20). However, if Neandertals are, onaverage across many independent regions of thegenome, more closely related to present-day hu-mans in certain parts of the world than in others,this would strongly suggest that Neandertals ex-

changed parts of their genome with the ances-tors of these groups.

Several features of DNA extracted from LatePleistocene remains make its study challenging.The DNA is invariably degraded to a small aver-age size of less than 200 base pairs (bp) (21, 22),it is chemically modified (21, 23–26), and extractsalmost always contain only small amounts of en-dogenous DNA but large amounts of DNA frommicrobial organisms that colonized the specimensafter death. Over the past 20 years, methods forancientDNAretrieval have been developed (21,22),largely based on the polymerase chain reaction(PCR) (27). In the case of the nuclear genome ofNeandertals, four short gene sequences have beendetermined by PCR: fragments of theMC1R geneinvolved in skin pigmentation (28), a segment ofthe FOXP2 gene involved in speech and language(29), parts of the ABO blood group locus (30), anda taste receptor gene (31). However, although PCRof ancient DNA can be multiplexed (32), it doesnot allow the retrieval of a large proportion of thegenome of an organism.

The development of high-throughput DNA se-quencing technologies (33, 34) allows large-scale,genome-wide sequencing of random pieces ofDNA extracted from ancient specimens (35–37)and has recently made it feasible to sequence ge-

RESEARCHARTICLE

1Department of Evolutionary Genetics, Max-Planck Institute forEvolutionary Anthropology, D-04103 Leipzig, Germany. 2BroadInstitute of MIT and Harvard, Cambridge, MA 02142, USA.3Department of Integrative Biology, University of California,Berkeley, CA 94720, USA. 4European Molecular BiologyLaboratory–European Bioinformatics Institute, Wellcome TrustGenome Campus, Hinxton, Cambridgeshire, CB10 1SD, UK.5Genome Technology Branch, National Human Genome Re-search Institute, National Institutes of Health, Bethesda, MD20892, USA. 6Program in Bioinformatics and Integrative Biology,University of Massachusetts Medical School, Worcester, MA01655, USA. 7Howard Hughes Medical Institute, Departmentof Genome Sciences, University of Washington, Seattle, WA98195, USA. 8Division of Biological Sciences, University ofMontana, Missoula, MT 59812, USA. 9454 Life Sciences,Branford, CT 06405, USA. 10Croatian Academy of Sciences andArts, Zrinski trg 11, HR-10000 Zagreb, Croatia. 11CroatianAcademy of Sciences and Arts, Institute for QuaternaryPaleontology andGeology, Ante Kovacica 5, HR-10000 Zagreb,Croatia. 12ANO Laboratory of Prehistory, St. Petersburg, Russia.13Institute of Evolutionary Biology (UPF-CSIC), Dr. Aiguader88, 08003 Barcelona, Spain. 14Área de Prehistoria Departa-mento de Historia Universidad de Oviedo, Oviedo, Spain.15Departamento de Paleobiología, Museo Nacional de CienciasNaturales, CSIC, Madrid, Spain. 16Der LandschaftverbandRheinlund–Landesmuseum Bonn, Bachstrasse 5-9, D-53115Bonn, Germany. 17Abteilung für Vor- und FrühgeschichtlicheArchäologie, Universität Bonn, Germany. 18Department ofBiology, EmoryUniversity, Atlanta, GA 30322,USA. 19Departmentof Microbiology, University College Cork, Cork, Ireland. 20Depart-ment of Genetics, Harvard Medical School, Boston, MA 02115,USA. 21Department of Human Evolution, Max-Planck Institutefor Evolutionary Anthropology, D-04103 Leipzig, Germany.

*To whom correspondence should be addressed. E-mail:[email protected] (R.E.G.); [email protected] (D.R.); [email protected] (S.P.)†Members of the Neandertal Genome Analysis Consortium.‡Present address: Department of Biomolecular Engineer-ing, University of California, Santa Cruz, CA 95064, USA.§These authors contributed equally to this work.||Present address: Beijing Institute of Genomics, ChineseAcademy of Sciences Beijing 100029, P.R. China.¶Deceased.

7 MAY 2010 VOL 328 SCIENCE www.sciencemag.org710

on

Mar

ch 2

4, 2

013

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fro

m

2-4% of eurasian DNA comes from Neanderthals


Strong reproductive isolation between humansand Neanderthals inferred from observedpatterns of introgressionMathias Currata,1 and Laurent Excoffierb,c,1

aAnthropology, Genetics, and Peopling History Laboratory, Anthropology Unit, Department of Genetics and Evolution, University of Geneva,1227 Geneva, Switzerland; bComputational and Molecular Population Genetics Laboratory, Institute of Ecology and Evolution, University of Berne,3012 Berne, Switzerland; and cSwiss Institute of Bioinformatics, 1015 Lausanne, Switzerland

Edited by Svante Pääbo, Max Planck Institute of Evolutionary Anthropology, Leipzig, Germany, and approved August 3, 2011 (received for reviewMay 10, 2011)

Recent studies have revealed that 2–3% of the genome of non-Africansmight come fromNeanderthals, suggesting amore complexscenario of modern human evolution than previously anticipated. Inthis paper, we use a model of admixture during a spatial expansionto study the hybridization of Neanderthals with modern humansduring their spread out of Africa. We find that observed low levelsof Neanderthal ancestry in Eurasians are compatible with a very lowrate of interbreeding (<2%), potentially attributable to a very strongavoidance of interspecific matings, a low fitness of hybrids, or both.These results suggesting the presence of very effective barriers togeneflowbetween the twospecies are robust touncertainties aboutthe exact demography of the Paleolithic populations, and they arealso found to be compatible with the observed lack of mtDNA in-trogression.Ourmodel additionally suggests that similarly low levelsof introgression in Europe and Asia may result from distinct admix-ture events havingoccurredbeyond theMiddle East, after the split ofEuropeans and Asians. This hypothesis could be tested because itpredicts that different components of Neanderthal ancestry shouldbe present in Europeans and in Asians.

genetic introgression | simulation

Recent analyses have revealed that Neanderthal genomes showcloser genetic affinities with contemporary non-Africans than

with West Africans (1, 2). Although this could be attributable tothe existence of ancient subdivisions within Africa, it seems betterexplained by ancient episodes of admixture between Neanderthalsand early Eurasians (1, 2). These results are at odds with thoseobtained frommtDNA (3–5) and challenge the simplest version ofan out-of-Africa model of human evolution, which posits a com-plete replacement of Neanderthals by modern humans (e.g., 6, 7).It is thus likely that anatomically modern humans (Homo sapiens)have hybridized with Neanderthals (1), Denisovans (2), and po-tentially other archaic humans (e.g., 8, 9) while migrating out ofAfrica. However, the current quantification of the introgressiondoes not shedmuch light on the hybridization process itself, whichremains relatively unclear.For instance, similar levels of Neanderthal introgression are

observed in Europe and in Asia (1), which has been interpreted asevidence for a single and limited episode of admixture betweenNeanderthals and the ancestors of Eurasians some 50–60 kya (1).This interpretation implies either that there has been no sub-sequent admixture between modern humans and Neanderthalswhen the formers colonized Europe some 40 kya (10–12) or thatsome admixture occurred in Europe, where these species coex-isted (13), but that this signal has now disappeared because of drift(14) or latermigrations of nonadmixed populations (1).Moreover,the very low level of Neanderthal ancestry observed in Eurasians(1.9–3.1%) (2) is somehow surprising, because one would expectto see massive levels of Neanderthal introgression into modernhumans if admixture was not strongly prevented during the rangeexpansion of modern humans out of Africa (3, 15).

To examine these issues and clarify the process of hybridizationbetween Neanderthals and modern humans, we have used a real-istic and spatially explicit model of admixture and competitionbetween modern humans and Neanderthals (3). Using extensivesimulations, we have estimated the interbreeding success ratebetween humans and Neanderthals as well as the spatial scale ofhybridization that is compatible with the observed patterns ofNeanderthal ancestry in contemporary humans, assuming that thelatter migrated out of Africa into Eurasia 50 kya (6, 7).

ResultsLow Rates of Interbreeding Between Humans and Neanderthals.Using spatially explicit simulations, we have computed theexpected amount of Neanderthal ancestry in present-day samplesfrom Europe (France) and Asia (China) for different levels ofadmixture with Neanderthals and over various possible Nean-derthal ranges (Fig. 1). Under our model of admixture duringrange expansion, we find that observed low levels of Neanderthalintrogression into Eurasians imply the existence of extremelystrong barriers to gene flow between the two species (Table 1)because of a very low fitness of human-Neanderthal hybrids, a verystrong avoidance of interspecificmating, or a combination of thesepre- and postzygotic barriers. Indeed, under most investigateddemographic scenarios, the interbreeding success rate betweenhumans and Neanderthals was found to be below 2% (Fig. 2,Table 1, Fig. S1, and Table S1). Under demographic scenario A(Table 1), which is based on the most realistic demographicparameters, we estimate this interbreeding success to be even wellbelow 1% (0.51%, mode of black curve in Fig. 2; 95% confidenceinterval: 0.33–0.89%). Higher estimates of interbreeding successare obtained by assuming lower local population densities (sce-narios C and C9), that hybridization between the two species onlyoccurred in a small area of the Middle East (1.5%, scenario A99;Fig. 2 and Table 1), or that population densities were higher in theMiddle East than in other regions of Europe or Asia (scenarios Gand G9). Contrastingly, lower interbreeding success estimates areobtained under demographic scenarios where the two species caninteract for a longer time, for example, where local populationdensities are higher (scenario B, 0.37%) or where populationgrowth is slower (scenario E, 0.30%). As can be seen in Fig. 2,interbreeding success generally decreases with increasing pop-ulation densities and with decreasing expansion speeds, but it is

Author contributions: M.C. and L.E. designed research; M.C. performed research; M.C. andL.E. analyzed data; and M.C. and L.E. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.1To whom correspondence may be addressed. E-mail: [email protected] or [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1107450108/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1107450108 PNAS | September 13, 2011 | vol. 108 | no. 37 | 15129e15134

ANTH

ROPO

LOGY

Strong reproductive isolation between humansand Neanderthals inferred from observedpatterns of introgressionMathias Currata,1 and Laurent Excoffierb,c,1

aAnthropology, Genetics, and Peopling History Laboratory, Anthropology Unit, Department of Genetics and Evolution, University of Geneva,1227 Geneva, Switzerland; bComputational and Molecular Population Genetics Laboratory, Institute of Ecology and Evolution, University of Berne,3012 Berne, Switzerland; and cSwiss Institute of Bioinformatics, 1015 Lausanne, Switzerland

Edited by Svante Pääbo, Max Planck Institute of Evolutionary Anthropology, Leipzig, Germany, and approved August 3, 2011 (received for reviewMay 10, 2011)

Recent studies have revealed that 2–3% of the genome of non-Africansmight come fromNeanderthals, suggesting amore complexscenario of modern human evolution than previously anticipated. Inthis paper, we use a model of admixture during a spatial expansionto study the hybridization of Neanderthals with modern humansduring their spread out of Africa. We find that observed low levelsof Neanderthal ancestry in Eurasians are compatible with a very lowrate of interbreeding (<2%), potentially attributable to a very strongavoidance of interspecific matings, a low fitness of hybrids, or both.These results suggesting the presence of very effective barriers togeneflowbetween the twospecies are robust touncertainties aboutthe exact demography of the Paleolithic populations, and they arealso found to be compatible with the observed lack of mtDNA in-trogression.Ourmodel additionally suggests that similarly low levelsof introgression in Europe and Asia may result from distinct admix-ture events havingoccurredbeyond theMiddle East, after the split ofEuropeans and Asians. This hypothesis could be tested because itpredicts that different components of Neanderthal ancestry shouldbe present in Europeans and in Asians.

genetic introgression | simulation

Recent analyses have revealed that Neanderthal genomes showcloser genetic affinities with contemporary non-Africans than

with West Africans (1, 2). Although this could be attributable tothe existence of ancient subdivisions within Africa, it seems betterexplained by ancient episodes of admixture between Neanderthalsand early Eurasians (1, 2). These results are at odds with thoseobtained frommtDNA (3–5) and challenge the simplest version ofan out-of-Africa model of human evolution, which posits a com-plete replacement of Neanderthals by modern humans (e.g., 6, 7).It is thus likely that anatomically modern humans (Homo sapiens)have hybridized with Neanderthals (1), Denisovans (2), and po-tentially other archaic humans (e.g., 8, 9) while migrating out ofAfrica. However, the current quantification of the introgressiondoes not shedmuch light on the hybridization process itself, whichremains relatively unclear.For instance, similar levels of Neanderthal introgression are

observed in Europe and in Asia (1), which has been interpreted asevidence for a single and limited episode of admixture betweenNeanderthals and the ancestors of Eurasians some 50–60 kya (1).This interpretation implies either that there has been no sub-sequent admixture between modern humans and Neanderthalswhen the formers colonized Europe some 40 kya (10–12) or thatsome admixture occurred in Europe, where these species coex-isted (13), but that this signal has now disappeared because of drift(14) or latermigrations of nonadmixed populations (1).Moreover,the very low level of Neanderthal ancestry observed in Eurasians(1.9–3.1%) (2) is somehow surprising, because one would expectto see massive levels of Neanderthal introgression into modernhumans if admixture was not strongly prevented during the rangeexpansion of modern humans out of Africa (3, 15).

To examine these issues and clarify the process of hybridizationbetween Neanderthals and modern humans, we have used a real-istic and spatially explicit model of admixture and competitionbetween modern humans and Neanderthals (3). Using extensivesimulations, we have estimated the interbreeding success ratebetween humans and Neanderthals as well as the spatial scale ofhybridization that is compatible with the observed patterns ofNeanderthal ancestry in contemporary humans, assuming that thelatter migrated out of Africa into Eurasia 50 kya (6, 7).

ResultsLow Rates of Interbreeding Between Humans and Neanderthals.Using spatially explicit simulations, we have computed theexpected amount of Neanderthal ancestry in present-day samplesfrom Europe (France) and Asia (China) for different levels ofadmixture with Neanderthals and over various possible Nean-derthal ranges (Fig. 1). Under our model of admixture duringrange expansion, we find that observed low levels of Neanderthalintrogression into Eurasians imply the existence of extremelystrong barriers to gene flow between the two species (Table 1)because of a very low fitness of human-Neanderthal hybrids, a verystrong avoidance of interspecificmating, or a combination of thesepre- and postzygotic barriers. Indeed, under most investigateddemographic scenarios, the interbreeding success rate betweenhumans and Neanderthals was found to be below 2% (Fig. 2,Table 1, Fig. S1, and Table S1). Under demographic scenario A(Table 1), which is based on the most realistic demographicparameters, we estimate this interbreeding success to be even wellbelow 1% (0.51%, mode of black curve in Fig. 2; 95% confidenceinterval: 0.33–0.89%). Higher estimates of interbreeding successare obtained by assuming lower local population densities (sce-narios C and C9), that hybridization between the two species onlyoccurred in a small area of the Middle East (1.5%, scenario A99;Fig. 2 and Table 1), or that population densities were higher in theMiddle East than in other regions of Europe or Asia (scenarios Gand G9). Contrastingly, lower interbreeding success estimates areobtained under demographic scenarios where the two species caninteract for a longer time, for example, where local populationdensities are higher (scenario B, 0.37%) or where populationgrowth is slower (scenario E, 0.30%). As can be seen in Fig. 2,interbreeding success generally decreases with increasing pop-ulation densities and with decreasing expansion speeds, but it is

Author contributions: M.C. and L.E. designed research; M.C. performed research; M.C. andL.E. analyzed data; and M.C. and L.E. wrote the paper.


This article is a PNAS Direct Submission.1To whom correspondence may be addressed. E-mail: [email protected] or [email protected].

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1107450108/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1107450108 PNAS | September 13, 2011 | vol. 108 | no. 37 | 15129e15134

ANTH

ROPO

LOGY


• Only known remains(all found since 2010): phalanx (finger bone), three teeth, a toe bone. From 41,000 years ago.

• Amazingly well preserved DNA (Siberia; average temperature 0°C). ⇒ sequenced the genome.

• Common ancestor with Neanderthal: 600,000 years ago

• Interbreeding with Homo sapiens: 4-6% of Melanesian genomes are from Denisovan.

Nature 2010ARTICLEdoi:10.1038/nature09710

Genetic history of an archaic hominingroup from Denisova Cave in SiberiaDavid Reich1,2*, Richard E. Green3,4*, Martin Kircher3*, Johannes Krause3,5*, Nick Patterson2*, Eric Y. Durand6*, Bence Viola3,7*,Adrian W. Briggs1,3, Udo Stenzel3, Philip L. F. Johnson8, Tomislav Maricic3, Jeffrey M. Good9, Tomas Marques-Bonet10,11,Can Alkan10, Qiaomei Fu3,12, Swapan Mallick1,2, Heng Li2, Matthias Meyer3, Evan E. Eichler10, Mark Stoneking3,Michael Richards7,13, Sahra Talamo7, Michael V. Shunkov14, Anatoli P. Derevianko14, Jean-Jacques Hublin7, Janet Kelso3,Montgomery Slatkin6 & Svante Paabo3

Using DNA extracted from a finger bone found in Denisova Cave in southern Siberia, we have sequenced the genome of anarchaic hominin to about 1.9-fold coverage. This individual is from a group that shares a common origin withNeanderthals. This population was not involved in the putative gene flow from Neanderthals into Eurasians; however,the data suggest that it contributed 4–6% of its genetic material to the genomes of present-day Melanesians. We designatethis hominin population ‘Denisovans’ and suggest that it may have been widespread in Asia during the Late Pleistoceneepoch. A tooth found in Denisova Cave carries a mitochondrial genome highly similar to that of the finger bone. This toothshares no derived morphological features with Neanderthals or modern humans, further indicating that Denisovanshave an evolutionary history distinct from Neanderthals and modern humans.

Less than 200,000 years ago, anatomically modern humans (that is,humans with skeletons similar to those of present-day humans)appeared in Africa. At that time, as well as later when modern humansappeared in Eurasia, other ‘archaic’ hominins were already present inEurasia. In Europe and western Asia, hominins defined as Neanderthalson the basis of their skeletal morphology lived from at least 230,000years ago before disappearing from the fossil record about 30,000 yearsago1. In eastern Asia, no consensus exists about which groups werepresent. For example, in China, some have emphasized morphologicalaffinities between Neanderthals and the specimen of Maba2, or betweenHomo heidelbergensis and the Dali skull3. However, others classify thesespecimens as ‘early Homo sapiens’4. In addition, until at least 17,000years ago, Homo floresiensis, a short-statured hominin that seems torepresent an early divergence from the lineage leading to present-dayhumans5–7, was present on the island of Flores in Indonesia and possiblyelsewhere.

DNA sequences retrieved from hominin remains offer an approachcomplementary to morphology for understanding hominin relation-ships. For Neanderthals, the nuclear genome was recently determinedto about 1.3-fold coverage8. This revealed that Neanderthal DNAsequences and those of present-day humans share common ancestorson average about 800,000 years ago and that the population split ofNeanderthal and modern human ancestors occurred 270,000–440,000 years ago. It also showed that Neanderthals shared moregenetic variants with present-day humans in Eurasia than with pre-sent-day humans in sub-Saharan Africa, indicating that gene flowfrom Neanderthals into the ancestors of non-Africans occurred toan extent that 1–4% of the genomes of people outside Africa arederived from Neanderthals8. In addition, ten partial and six complete

mitochondrial (mt)DNA sequences have been determined fromNeanderthals9–17. This has shown that all Neanderthals studied sofar share a common mtDNA ancestor on the order of 100,000 yearsago10, and in turn, share a common ancestor with the mtDNAs ofpresent-day humans about 500,000 years ago10,18,19 (as expected, this isolder than the Neanderthal–modern human population split time of270,000–440,000 years ago estimated from the nuclear genome8). Oneof these mtDNA sequences has also shown that hominins carryingmtDNAs typical of Neanderthals were present as far east as the AltaiMountains in southern Siberia13.

In 2008, the distal manual phalanx of a juvenile hominin was exca-vated at Denisova Cave. This site is located in the Altai Mountains insouthern Siberia, and is a reference site for the Middle to UpperPalaeolithic of the region where systematic excavations over the past25 years have uncovered cultural layers indicating that human occu-pation at the site started up to 280,000 years ago20. The phalanx wasfound in layer 11, which has been dated to 50,000 to 30,000 years ago.This layer contains microblades and body ornaments of polishedstone typical of the ‘Upper Palaeolithic industry’ generally thoughtto be associated with modern humans, but also stone tools that aremore characteristic of the earlier Middle Palaeolithic, such as side-scrapers and Levallois blanks21–23.

Recently, we used a DNA capture approach10 in combination withhigh-throughput sequencing to determine a complete mtDNA genomefrom the Denisova phalanx. Surprisingly, this mtDNA diverged fromthe common lineage leading to modern human and NeanderthalmtDNAs about one million years ago19, that is, about twice as far backin time as the divergence between Neanderthal and modern humanmtDNAs. However, mtDNA is maternally inherited as a single unit

*These authors contributed equally to this work.

1Department of Genetics, Harvard Medical School, Boston, Massachusetts 02115, USA. 2Broad Institute of MIT and Harvard, Cambridge, Massachusetts 02142, USA. 3Department of Evolutionary Genetics,Max Planck Institute for Evolutionary Anthropology, Leipzig 04103, Germany. 4Department of Biomolecular Engineering, University of California, Santa Cruz 95064, USA. 5Institut furNaturwissenschaftliche Archaologie, University of Tubingen, Tubingen 72070, Germany. 6Department of Integrative Biology, University of California, Berkeley, California 94720, USA. 7Department ofHuman Evolution, Max Planck Institute for Evolutionary Anthropology, Leipzig 04103, Germany. 8Department of Biology, Emory University, Atlanta, Georgia 30322, USA. 9Division of Biological Sciences,University of Montana, Missoula, Montana 59812, USA. 10Howard Hughes Medical Institute, Department of Genome Sciences, University of Washington, Seattle, Washington 98195, USA. 11Institute ofEvolutionary Biology (UPF-CSIC), 08003 Barcelona, Spain. 12CAS-MPS Joint Laboratory for Human Evolution and Archeometry, Institute of Vertebrate Paleontology and Paleoanthropology of ChineseAcademy of Sciences, Beijing 100044, China. 13Department of Anthropology, University of British Columbia, Vancouver, British Columbia V6T 1Z1, Canada. 14Palaeolithic Department, Institute ofArchaeology & Ethnography, Russian Academy of Sciences, Siberian Branch, Novosibirsk 630090, Russia.

2 3 / 3 0 D E C E M B E R 2 0 1 0 | V O L 4 6 8 | N A T U R E | 1 0 5 3

Macmillan Publishers Limited. All rights reserved©2010



Proconsulidae

Australopithecines

H. erectus

H. neanderthalensis

H. sapiens

Denisovan


1WV�QH�#HTKEC`��[GCTU�CIQ

0Q�&GPKUQXCPCFOKZVWTG

&GPKUQXCPCFOKZVWTG

!

&GPKUQXC�%CXG

�

�

!

0GCPFGTVJCNCFOKZVWTG

0CVWTG�4GXKGYU�^�)GPGVKEUFigure 4 | Dispersal of modern humans from Africa. A map illustrating the FKURGTUCN�QH�OQFGTP�JWOCPU�HTQO�#HTKEC�CDQWV��|[GCTU�CIQ��HQNNQYGF� by admixture with Neanderthals in the ancestry of all non-Africans, followed by admixture with Denisovans in the ancestry of New Guineans. Arrows indicate general directionality and not specific migration routes — in general we only know for sure the end points of migrations, not the routes. The red star indicates the location of Denisova Cave. The exclamation marks indicate admixture, but there is extreme uncertainty as to where the Neanderthal and Denisovan admixture occurred. Question marks indicate regions where no additional admixture was detected even though archaeological findings suggest that Neanderthals and Denisovans overlapped with modern human populations in those regions.

with a genomic divergence similar to, or slightly larger than, the deepest divergence known among present-day humans, that between San and other groups3,6. After Neanderthals and Denisovans diverged they experi-enced independent histories, which are reflected in their genetic contributions to different present-day human groups. Such information was not apparent from the morphology of the meagre fossil remains attributed to Denisovans. Identifying the source of a particular fos-sil in the absence of any informative morphology, and even identifying previously unknown hominin groups, as in the case of Denisovans, is likely to be a powerful application of ancient DNA in the future.

It should be stressed that genome sequence diver-gence times are older than population divergence times because of genetic polymorphism in the ancestral pop-ulations61. That is, if at the time of population diver-gence there is polymorphism at a nucleotide site within the ancestral population, then the genetic divergence at that nucleotide site must be older than the popula-tion divergence time. However, with some assumptions about population history, genome sequence data can be used to estimate population divergence times, and the resulting estimate for the divergence of the ances-tors of Neanderthals and Denisovans from the ancestors of modern humans is about 350 kya1,3. This presum-ably reflects the time when a hominin population left Africa and evolved into Neanderthals and Denisovans, while other hominins in Africa evolved into modern humans.

African origin of modern humans. Single-locus studies of mtDNA and NRY variation in modern human popu-lations have strongly supported a recent African origin of our species, in terms of Africa being the source of the deepest lineages and harbouring the greatest diver-sity14,62–65. Genome-wide SNP data is consistent with this view7–9, and genome sequences from several modern humans indicate that the deepest population divergences within modern humans are between San individuals from southern Africa and other groups, approxi-mately 115 kya1,6. Genetic data indicate that modern humans first dispersed from Africa about 50 kya with divergences among non-African populations dating to 35–50 kya13,66–68. One of the most convincing indications of a strong signal of a recent African origin throughout our genome was the demonstration of an astonishingly close correlation between the amount of genetic diversity in a population and the geographic distance of that pop-ulation from East Africa69. This ‘serial bottleneck’ model strongly implies an African origin of modern humans; in summary, the genetic evidence for an African origin of modern humans is overwhelming.

Dispersal from Africa: replacement or assimilation? Given that modern humans arose recently in Africa and given that other hominins (such as Neanderthals and Denisovans) already existed outside Africa, what happened when modern humans dispersed from Africa and encountered these other hominins? Was there interbreeding, thereby leading to genetic contributions to modern humans from these non-African hominins, or were the non-African hominins replaced without any interbreeding? The extent to which non-African homi-nins might have contributed to the genomes of modern humans has been one of the long-standing controversies in human evolution70. In our opinion this can now be laid to rest, thanks to the Neanderthal and Denisovan genome sequences1,3: all non-Africans (and no sub-Saharan Africans) examined to date show about the same amount of gene flow from Neanderthals, with an estimated 2–4% of the genomes of non-Africans com-ing from Neanderthals. It is also possible to explain this signal of Neanderthal gene flow by a more complicated scenario involving deep population structure within Africa1. However, the finding of a signal of gene flow into some modern humans from the Denisova homi-nin renders this alternative explanation less likely3, and in our opinion the model that best explains human origins is a recent African origin followed by a small amount of admixture (or assimilation) with non-African hominins (FIG. 4).

Dispersal from Africa: how many times, and which way did they go? The Neanderthal and Denisovan genome sequences give us new insights into human migrations, as the presence (or absence) of the signal of a genetic contribution from a particular extinct hominin can be used as a marker of population relationships. Whether there was a single dispersal or multiple dispersals of modern humans from Africa has been a long-standing question67,71. The finding that all modern non-Africans

REVIEWS

NATURE REVIEWS | GENETICS VOLUME 12 | SEPTEMBER 2011 | 611

© 2011 Macmillan Publishers Limited. All rights reserved

Stoneking & Krause 2011

Stoneking & Krause 2011 Nature Reviews Genetics

? No additional admixture detected despite probable overlap! detected admixture (location uncertain)


PATCHWORK PLANET

2%

98%

Most people’s genomes contain remnants of archaic DNA from ancient interbreeding3–6.

Genes*AfricanUnknown archaicAfrican sourceNeanderthalDenisovan

2.5% 5%2.5%

97.5% 92.5%

*Figures are approximate, and for Africa, based on

A WINDING PATHAfter early modern humans left Africa around 60,000 years ago (top right), they spread across the globe and interbred with other descendants of Homo heidelbergensis.

Homo sapiens

2.0

1.6

1.2

0.8

Mill

ion

year

s ag

o

0.4

0

Homo erectus

Homo antecessor

Homo !oresiensis Denisovans Neanderthals

Homo heidelbergensis

Homo erectus

Wavy branch edges suggest presumed !uctuations in population.

H. !oresiensis originated in an unknown location and reached remote parts of Indonesia.

H. erectus spread to western Asia, then east Asia and Indonesia. Its presence in Europe is uncertain, but it gave rise to H. antecessor, found in Spain.

H. heidelbergensis originated from H. erectus in an unknown location and dispersed across Africa, southern Asia and southern Europe.

H. sapiens spread from Africa to western Asia and then to Europe and southern Asia, eventually reaching Australasia and the Americas.

Stringer 2012 NatureMonday, 14 October 13


What about today?

Does natural selection (still) act on humans?


But... Triple (?) misunderstandings:

1. Evolution (ie genetic change) is not only through natural selection

•Drift • Sexual selection• ...

2. Medicine can reduce effect of deleterious traits. •OK many are ‘‘alive who otherwise would have perished’’•But many have limited access to medicine.

3. Even with the best medical care• there are differences in reproductive success


Some examples


Pathogens/disease

• “Spanish” Flu pandemic of 1918 killed 1-3% of the world’s population




Pathogens/disease

• “Spanish” Flu pandemic of 1918 killed 1-3% of the world’s population

• AIDS

• Dengue, Typhus, Malaria...


Other examples


Genetic Evidence for High-AltitudeAdaptation in TibetTatum S. Simonson,1 Yingzhong Yang,2* Chad D. Huff,1 Haixia Yun,2* Ga Qin,2*David J. Witherspoon,1 Zhenzhong Bai,2* Felipe R. Lorenzo,3 Jinchuan Xing,1Lynn B. Jorde,1† Josef T. Prchal,1,3† RiLi Ge2*†

Tibetans have lived at very high altitudes for thousands of years, and they have a distinctivesuite of physiological traits that enable them to tolerate environmental hypoxia. These phenotypesare clearly the result of adaptation to this environment, but their genetic basis remainsunknown. We report genome-wide scans that reveal positive selection in several regions thatcontain genes whose products are likely involved in high-altitude adaptation. Positively selectedhaplotypes of EGLN1 and PPARA were significantly associated with the decreased hemoglobinphenotype that is unique to this highland population. Identification of these genes providessupport for previously hypothesized mechanisms of high-altitude adaptation and illuminates thecomplexity of hypoxia-response pathways in humans.

The Tibetan highlands are one of the mostextreme environments inhabited by hu-mans. Many present-day Tibetan popula-

tions are thought to be descendants of peoplewho have occupied the Tibetan Plateau sincethe mid-Holocene, between 7000 and 5000 yearsago (1), and possibly since the late Pleistocene,~21,000 years ago (2, 3). Compared with Andeanpopulations living in similar high-altitude condi-tions, Tibetans exhibit a distinct suite of phys-iologic traits: decreased arterial oxygen content,increased resting ventilation, lack of hypoxic pul-monary vasoconstriction, lower incidence of re-duced birth weight, and reduced hemoglobin (Hb)concentration (on average, 3.6 g/dl less for bothmales and females) (4–8). Neighboring HanChinese individuals and other nonadapted lowlandvisitors to high-altitude regions develop increasedHb concentration to compensate for the hypoxichigh-altitude environment (9), and this responseis associated with adverse effects (10, 11).

High-altitude Tibetans maintain normal aero-bic metabolism, despite profound arterial hypoxia(4), perhaps through the existence of changes inthe oxygen-transport system. For example, ele-vated circulating NO levels increase vasodilationand blood flow (12), which, when combined withincreased ventilation (13), may increase the avail-ability of oxygen to cells (4). Collectively, thesetraits strongly suggest that Tibetans have adapteduniquely to extreme high-altitude conditions. Thegenetic basis of this adaptation, however, remainsunknown.

We used two intersecting criteria to identifygenes potentially involved in high-altitude adap-tation: First, a priori candidates for adaptation tohigh-altitude hypoxia were chosen because oftheir known functions (14). Second, a genome-wide scan was conducted to identify regions thatshow strong evidence of local positive selectionin high-altitude Tibetans (Fig. 1). To generate a

set of a priori functional candidate loci, we con-structed a list of Gene Ontology (GO) projectcategories (15) associated with the traits discussedabove (Table 1). We merged genes from this listwith those in the Panther-defined pathway “hy-poxia response via activation of hypoxia-induciblefactor (HIF)” (16), a major transcriptional regu-lator of oxygen homeostasis (17) that is prob-ably associated with high-altitude adaptation. Theresulting set of 247 functional candidate loci islisted in table S2.

We next identified alleles subject to strongrecent positive selection (a selective sweep) in asample of 31 unrelated Tibetans who were geno-typed for one million single-nucleotide polymor-phisms (SNPs) using the Affymetrix Genome-WideHuman SNP 6.0 Array. These individuals showedno evidence of admixture with neighboring pop-ulations [see supporting online material (SOM),figs. S1 and S2]. To pinpoint loci under positiveselection, we first used the cross-population ex-tended haplotype homozygosity (XP-EHH) sta-tistic (18) to make comparisons between theTibetan highland population and the combinedHapMap Chinese (CHB) and Japanese (JPT)lowland populations (19). The XP-EHH statisticassesses haplotype differences between two pop-ulations and is designed to detect alleles that

1Eccles Institute of Human Genetics, University of Utah Schoolof Medicine, Salt Lake City, UT 84112, USA. 2Research Centerfor High-Altitude Medicine, Qinghai University MedicalSchool, Xining, Qinghai 810001, People’s Republic of China.3Division of Hematology and Department of Pathology (ARUP),University of Utah School of Medicine and VAH, Salt Lake City,UT 84112, USA.

*The Research Center for High-Altitude Medicine initiated theresearch project and was primarily responsible for phenotyp-ing and DNA collection.†To whom correspondence should be addressed. E-mail:[email protected] (L.B.J); [email protected](J.T.P.); [email protected] (R.L.G.)

Fig. 1. Gene regions respon-sible for adaptation to high-altitude hypoxia in Tibetans.(A) The strategy used to iden-tify a list of genes related tohigh-altitude adaptation tohypoxia relies on three setsof genes. The set of function-al candidates (yellow) consistsof genes associated with phys-iological traits related tohypoxia (see Table 1 for cate-gories). The XP-EHH (lightblue) and iHS (dark blue)selection candidate sets in-clude genes in the top 1%of the empirical distribu-tions of XP-EHH and iHS re-sults, respectively, excludingthose with evidence of posi-tive selection in neighboringpopulations (see SOM). Theintersection of functional can-didates with selection can-didates (outlined in black) isenriched for regions contain-ing genes that contribute tolocal adaptation to hypoxia in Tibetans. The genes in the intersection of functional candidates with iHSselection candidates still exhibit genetic variability in the population. (B to D) Comparison betweenTibetan and CHB-JPT genomic regions identified in selection scans. The top and bottom halves of eachfigure represent chromosome regions in the Tibetan (number of chromosomes = 62) and CHB-JPTpopulations (62 randomly drawn chromosomes from 90 individuals), respectively, for the (B) EPAS1, (C)EGLN1, and (D) HMOX2 genes identified in XPEHH, both scans, and iHS, respectively. The three SNPswith the highest iHS and XP-EHH scores (indicated by an asterisk) were designated as the corehaplotype for each genomic region. All haplotypes were sorted to the horizontal midline of each panelbased on the length of uninterrupted matches to the reference sequence. See fig. S3 and table S5 forthe remaining seven regions and details about these regions.

EPAS1

*

*

B

EGLN1 HMOX2D

*

iHScandidates

9919710

15 4

Functionalcandidates

237

XP-EHHcandidates

C

A

2 JULY 2010 VOL 329 SCIENCE www.sciencemag.org72

REPORTS

Genetic Evidence for High-AltitudeAdaptation in TibetTatum S. Simonson,1 Yingzhong Yang,2* Chad D. Huff,1 Haixia Yun,2* Ga Qin,2*David J. Witherspoon,1 Zhenzhong Bai,2* Felipe R. Lorenzo,3 Jinchuan Xing,1Lynn B. Jorde,1† Josef T. Prchal,1,3† RiLi Ge2*†

Tibetans have lived at very high altitudes for thousands of years, and they have a distinctivesuite of physiological traits that enable them to tolerate environmental hypoxia. These phenotypesare clearly the result of adaptation to this environment, but their genetic basis remainsunknown. We report genome-wide scans that reveal positive selection in several regions thatcontain genes whose products are likely involved in high-altitude adaptation. Positively selectedhaplotypes of EGLN1 and PPARA were significantly associated with the decreased hemoglobinphenotype that is unique to this highland population. Identification of these genes providessupport for previously hypothesized mechanisms of high-altitude adaptation and illuminates thecomplexity of hypoxia-response pathways in humans.

The Tibetan highlands are one of the mostextreme environments inhabited by hu-mans. Many present-day Tibetan popula-

tions are thought to be descendants of peoplewho have occupied the Tibetan Plateau sincethe mid-Holocene, between 7000 and 5000 yearsago (1), and possibly since the late Pleistocene,~21,000 years ago (2, 3). Compared with Andeanpopulations living in similar high-altitude condi-tions, Tibetans exhibit a distinct suite of phys-iologic traits: decreased arterial oxygen content,increased resting ventilation, lack of hypoxic pul-monary vasoconstriction, lower incidence of re-duced birth weight, and reduced hemoglobin (Hb)concentration (on average, 3.6 g/dl less for bothmales and females) (4–8). Neighboring HanChinese individuals and other nonadapted lowlandvisitors to high-altitude regions develop increasedHb concentration to compensate for the hypoxichigh-altitude environment (9), and this responseis associated with adverse effects (10, 11).

High-altitude Tibetans maintain normal aero-bic metabolism, despite profound arterial hypoxia(4), perhaps through the existence of changes inthe oxygen-transport system. For example, ele-vated circulating NO levels increase vasodilationand blood flow (12), which, when combined withincreased ventilation (13), may increase the avail-ability of oxygen to cells (4). Collectively, thesetraits strongly suggest that Tibetans have adapteduniquely to extreme high-altitude conditions. Thegenetic basis of this adaptation, however, remainsunknown.

We used two intersecting criteria to identifygenes potentially involved in high-altitude adap-tation: First, a priori candidates for adaptation tohigh-altitude hypoxia were chosen because oftheir known functions (14). Second, a genome-wide scan was conducted to identify regions thatshow strong evidence of local positive selectionin high-altitude Tibetans (Fig. 1). To generate a

set of a priori functional candidate loci, we con-structed a list of Gene Ontology (GO) projectcategories (15) associated with the traits discussedabove (Table 1). We merged genes from this listwith those in the Panther-defined pathway “hy-poxia response via activation of hypoxia-induciblefactor (HIF)” (16), a major transcriptional regu-lator of oxygen homeostasis (17) that is prob-ably associated with high-altitude adaptation. Theresulting set of 247 functional candidate loci islisted in table S2.

We next identified alleles subject to strongrecent positive selection (a selective sweep) in asample of 31 unrelated Tibetans who were geno-typed for one million single-nucleotide polymor-phisms (SNPs) using the Affymetrix Genome-WideHuman SNP 6.0 Array. These individuals showedno evidence of admixture with neighboring pop-ulations [see supporting online material (SOM),figs. S1 and S2]. To pinpoint loci under positiveselection, we first used the cross-population ex-tended haplotype homozygosity (XP-EHH) sta-tistic (18) to make comparisons between theTibetan highland population and the combinedHapMap Chinese (CHB) and Japanese (JPT)lowland populations (19). The XP-EHH statisticassesses haplotype differences between two pop-ulations and is designed to detect alleles that

1Eccles Institute of Human Genetics, University of Utah Schoolof Medicine, Salt Lake City, UT 84112, USA. 2Research Centerfor High-Altitude Medicine, Qinghai University MedicalSchool, Xining, Qinghai 810001, People’s Republic of China.3Division of Hematology and Department of Pathology (ARUP),University of Utah School of Medicine and VAH, Salt Lake City,UT 84112, USA.

*The Research Center for High-Altitude Medicine initiated theresearch project and was primarily responsible for phenotyp-ing and DNA collection.†To whom correspondence should be addressed. E-mail:[email protected] (L.B.J); [email protected](J.T.P.); [email protected] (R.L.G.)

Fig. 1. Gene regions respon-sible for adaptation to high-altitude hypoxia in Tibetans.(A) The strategy used to iden-tify a list of genes related tohigh-altitude adaptation tohypoxia relies on three setsof genes. The set of function-al candidates (yellow) consistsof genes associated with phys-iological traits related tohypoxia (see Table 1 for cate-gories). The XP-EHH (lightblue) and iHS (dark blue)selection candidate sets in-clude genes in the top 1%of the empirical distribu-tions of XP-EHH and iHS re-sults, respectively, excludingthose with evidence of posi-tive selection in neighboringpopulations (see SOM). Theintersection of functional can-didates with selection can-didates (outlined in black) isenriched for regions contain-ing genes that contribute tolocal adaptation to hypoxia in Tibetans. The genes in the intersection of functional candidates with iHSselection candidates still exhibit genetic variability in the population. (B to D) Comparison betweenTibetan and CHB-JPT genomic regions identified in selection scans. The top and bottom halves of eachfigure represent chromosome regions in the Tibetan (number of chromosomes = 62) and CHB-JPTpopulations (62 randomly drawn chromosomes from 90 individuals), respectively, for the (B) EPAS1, (C)EGLN1, and (D) HMOX2 genes identified in XPEHH, both scans, and iHS, respectively. The three SNPswith the highest iHS and XP-EHH scores (indicated by an asterisk) were designated as the corehaplotype for each genomic region. All haplotypes were sorted to the horizontal midline of each panelbased on the length of uninterrupted matches to the reference sequence. See fig. S3 and table S5 forthe remaining seven regions and details about these regions.

EPAS1

*

*

B

EGLN1 HMOX2D

*

iHScandidates

9919710

15 4

Functionalcandidates

237

XP-EHHcandidates

C

A

2 JULY 2010 VOL 329 SCIENCE www.sciencemag.org72

REPORTS


Deep Human Genealogies Reveal aSelective Advantage to Be on anExpanding Wave FrontClaudia Moreau,1 Claude Bhérer,1 Hélène Vézina,2 Michèle Jomphe,2

Damian Labuda,1,3* Laurent Excoffier1,4,5*

Since their origin, human populations have colonized the whole planet, but the demographicprocesses governing range expansions are mostly unknown. We analyzed the genealogy of morethan one million individuals resulting from a range expansion in Quebec between 1686 and 1960and reconstructed the spatial dynamics of the expansion. We find that a majority of the presentSaguenay Lac-Saint-Jean population can be traced back to ancestors having lived directly on orclose to the wave front. Ancestors located on the front contributed significantly more to the currentgene pool than those from the range core, likely due to a 20% larger effective fertility of womenon the wave front. This fitness component is heritable on the wave front and not in the core,implying that this life-history trait evolves during range expansions.

Most species go through environmental-ly induced range expansions or rangeshifts (1), promoting the evolution of

traits associated with dispersal and reproduction(2). Humans likely colonized the world by aseries of range expansions from Africa (3), pos-sibly with episodes of interbreeding with nowextinct hominins (4, 5), leading to allele frequen-cy and heterozygosity clines from entry pointsinto several continents [e.g., (6, 7)]. Range ex-pansions can also lead to drastic changes in allelefrequencies, sometimes mimicking the effect ofpositive selection in recently colonized habitats(8, 9), through a process called gene surfing (9).Neutral, favorable, or even deleterious mutationscan surf and increase in frequency (10, 11), im-plying that wave fronts may harbor mutationswith a wider range of selective coefficients thancore populations. The evolutionary consequencesof range expansions have been studied in a widearray of species (2, 12), but studies of the dy-namics of range expansions have been generallyrestricted to species with short generation times(13, 14) or to invasive species (15, 16), becauseboth spatial and temporal sampling are requiredto understand the dynamics of wave fronts.

Deep-rooted human genealogies in recentlyexpanded populations may offer an opportunityto study the wave front demographics and itsgenetic consequences on present-day populations.We studied the genealogies reconstructed from

Quebec parish registers that document the recenttemporal and spatial expansion of the settle-ment of the Charlevoix Saguenay Lac-Saint-Jean (ChSLSJ) region, northeast of Quebec City,Canada: a prime example of a recent, fast, andwell-documented range expansion (17) (Fig. 1).The European colonization of Quebec was ini-tiated in 1608 with the foundation of QuebecCity, and the colony was well established by theend of the 17th century (18). The peopling of theCharlevoix region started from Baie-Saint-Paul,and both a rapid demographic growth and the de-velopment of the timber industry promoted furtherexpansions after 1838 up the Saguenay River andthe Lac-Saint-Jean region (SLSJ) (19, 20). Thespatial and temporal dynamics of the peopling ofthe whole ChSLSJ region can be reconstructed by

tracing back the founding events of new localities.As shown in Fig. 1, the inferred colonization pro-cess is a mixture of long-distance settlementscreating an irregular wave front, followed by fur-ther, more progressive, short-range expansions,which then filled gaps and created a more reg-ular wave front.

On the basis of the computation of a wavefront index (WFI) (21), we find that the ancestorsof the Saguenay and the Lac-Saint-Jean peoplelived more often on or close to the wave frontthan expected by chance (WFI, P < 0.001 in bothregions) (fig. S1). Indeed, the very high WFI of0.75 observed in Lac-Saint-Jean corresponds toa situation in which half of the Lac-Saint-Jeanancestors had lived directly on the wave front andthe other half just one generation away from it.In contrast, WFI is significantly lower in theCharlevoix region (P = 0.003) (fig. S1). Theseresults are consistent with different colonizationdynamics of SLSJ and Charlevoix. The wavefront was always widespread in SLSJ where newlocalitieswere continuously settled, whereas it wasmuch smaller in Charlevoixwheremost localitiesremained in the range core until the 20th century(Fig. 1). New immigrants from outside ChSLSJconstituted an important minority of the peoplegetting married, with a greater proportion of im-migrants settling on the wave front than on therange core, especially before 1900 (up to 20% onthe wave front and up to 10% in the range core)(table S2). Generally, more male than female im-migration occurred in all regions, and this biastoward males is significantly higher in the corethan on the wave front (table S3). Nevertheless,the new territories of SLSJ have been largely col-onized by people recruited directly on the wavefront or next to it, not by people from the rangecore (table S4).

1686-1720

1721-1750

1751-1780

1781-1810

1811-1840

1841-1870

1871-1900

1901-1930

1931-1960

LacSaint-Jean

Saguenay River S

aint

Law

renc

e R

iver

Baie-Saint-Paul

Chicoutimi

Fig. 1. Map of the Charlevoix Saguenay Lac-Saint-Jean region showing the range expansion dynamicsand the wave front at different periods. Each filled circle represents a locality, and its color indicates itsage. Localities from the Charlevoix region are indicated by a black dot.

1Centre de Recherche, Hôpital Sainte-Justine, Université de Mon-tréal, 3175 Côte Sainte-Catherine, Montréal, Québec, Canada,H3T 1C5. 2Projet BALSAC, Université du Québec à Chicoutimi,555 Boulevard de l’Université, Chicoutimi, Québec, Canada, G7H2B1. 3Département de Pédiatrie, Université deMontréal, Montréal,Québec, Canada, H3T 1J4. 4Computational and Molecular Pop-ulation Genetics, Institute of Ecology and Evolution, Universityof Berne, Baltzerstrasse 6, 3012 Berne, Switzerland. 5SwissInstitute of Bioinformatics, 1015 Lausanne, Switzerland.

*To whom correspondence should be addressed. E-mail:[email protected] (L.E.); [email protected] (D.L.)

25 NOVEMBER 2011 VOL 334 SCIENCE www.sciencemag.org1148

REPORTS

on

Mar

ch 2

4, 2

013

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fro

m

We computed the expected number of genesleft by a given ancestor to the current genera-tion [its genetic contribution (GC)] (21) for allancestors of ChSLSJ, distinguishing betweenthose having reproduced on the wave front andthose in the range core (Table 1). We find thatover the entire studied period, individuals on thefront have contributed significantly more genesto the present generation than those in the core, inline with the theory predicting that surfing allelesshould be traced back to ancestors living on orclose to the wave front (22). We find similarresults when we restrict the analysis to the SLSJregion (Table 1), which has been colonized morerecently. Overall, ancestors on the edge contrib-uted 1.2 to 3.9 times more genes to the currentgeneration than ancestors from the core, the oldestancestors generally passing on more genes thanmore recent ones, in keeping with previous results[see, e.g., figure 4 in (23)]. In addition, 40.2% ofall ancestors of the ChSLSJ living between 1686and 1930 were on the wave front, reaching 45.1%for the SLSJ region (Table 1). For SLSJ, thenumber of ancestors living directly on the front orjust one generation away from it even reaches81% (table S4), showing the importance of thismoving edge for this region.

We compared the reproductive success ofwomen on the edge to the ones in the core, con-sidering both the number of their children [fami-ly size (FS)] and the number of their marriedchildren [effective family size (EFS)]. SLSJ fe-male ancestors living on the edge had on average15% more children than core SLSJ female an-cestors (Table 2) (P < 0.001) and 20% moremarried children (P < 0.001). These results showthat women’s fertility was significantly higher onthe wave front than in the range core and that thelarger genetic contribution of ancestors reproduc-ing on the wave front is likely not due purely to a

neutral surfing process but also to a net effect ofpositive selection on the front.

Women on the front overall had a slightlyhigher modal FS value (fig. S2A) and larger EFSdue to a right shift of the whole distribution to-ward higher values (fig. S2B), leading to a largerproportion of women on the front having morethan five married children (40% on the frontversus 26% in the core).We find only a slightlylower mortality rate of children under the age of5 on the wave front (23.6 versus 25.1% in thecore), implying that the increase in EFS comparedwith FS on the front is likely due to facilitatedaccess to reproduction (marriage). Interestingly,women on the front married almost 1 year earlierthan women in the core (Table 2), increasingtheir reproductive life, which may partly explaintheir overall higher fertility. This is in line withCharbonneau’s observations (18) that Quebecwomen had an overall longer reproductive lifecompared with French women of the same period,due to both an earlier age at first child and a laterage for their last child. However, our results sug-gest that the larger fertility of women in Quebecis mainly due to a front effect. An analysis ofcovariance (ANCOVA) reveals that the numberof children per women actually depends signif-icantly both on the age of marriage and on the

spatial location of reproduction (front or core)(P < 0.001 for the two effects) but that there is nointeraction between these factors (P = 0.46). Forthe number of married children, the two factors(P < 0.001) and their interaction (P = 0.02)are significant. We conclude that even thoughwomen on the front reproduce earlier thanwomen from the core, this contrast does not fullyexplain the difference in their fertility. The ad-vantage of being on the wave front remains if weuse less informative criteria to assign individualsto the front in SLSJ (tables S6 and S7), and it isnot due to a higher fertility of new immigrantssettling preferentially on the wave front (table S8).

We compared the fertility of women to theaverage fertility of their offspring (24), using thefact that the regression slope b gives us directly ameasure of heritability as h2 = 2b (25). Awomen’sFS is correlated with that of her own children onthe wave front, with a nonsignificantly differentheritability on the wave front (Fig. 2A) (h2 = 0.22)and in the range core (h2 = 0.16) (ANCOVA testof slope difference, P = 0.07). In contrast, if weconsider the transgenerational correlation in EFS,the correlation is only significant on the wavefront (Fig. 2B) (h2 = 0.24, ANCOVA test ofslope, P < 0.001) and not in the range core (h2 =0.04, ANCOVA test of slope, P = 0.23). The

Table 2. Age of reproduction and number of children of women from SLSJ in the period 1840 to 1900.Note that this table only includes women with known birth dates, such that age at marriage can becomputed.

No. ofwomen

Mean no. ofchildren(FS)

Mean no. ofmarriedchildren(EFS)

Mean age atmarriage

FS ratioWF/RC

EFS ratioWF/RC

Marriageage ratioWF/RC

Wave front (WF) 2663 9.1 4.9 20.51.15*** 1.20*** 0.95***

Range core (RC) 1783 7.9 4.1 21.6***, t test of difference between means; P < 0.001

Table 1. Genetic contribution (GC) of ancestors having lived in the ChSLSJ region to individuals from the 1931 to 1960 generation found anywhere in theQuebecprovince. Note that GCs of different generations are not independent.

Wave front (WF) Range core (RC)Generation

Total GCNo. of

ancestors ingenealogy

Mean GC Total GCNo. of

ancestors ingenealogy

Mean GCAncestors on wave

front (%)Mean GC

ratio (WF/RC)

ChSLSJ1686–1720 19298 48 402.0 612 6 102.0 88.9 3.94***1721–1750 19263 104 185.2 16833 106 158.8 49.5 1.17*1751–1780 22119 196 112.9 25990 373 69.7 34.4 1.62***1781–1810 21696 364 59.6 35613 1069 33.3 25.4 1.79***1811–1840 30504 1383 22.1 27061 1815 14.9 43.2 1.48***1841–1870 56589 6555 8.6 10175 2438 4.2 72.9 2.07***1871–1900 40386 8757 4.6 25619 8784 2.9 49.9 1.58***1901–1930 23370 10034 2.3 44408 26255 1.7 27.7 1.38***Total ChSLSJ 27441 40846 40.2

SLSJ1841–1870 27833 3743 7.4 39 15 2.6 99.6 2.8***1871–1900 33917 7300 4.6 15444 4420 3.5 62.3 1.3***1901–1930 21061 8832 2.4 35777 19726 1.8 30.9 1.3***Total SLSJ 19875 24161 45.1

*, P < 0.05 ***, P < 0.001

www.sciencemag.org SCIENCE VOL 334 25 NOVEMBER 2011 1149

REPORTS

on

Mar

ch 2

4, 2

013

ww

w.s

cien

cem

ag.o

rgD

ownl

oade

d fro

m



Natural selection in a contemporaryhuman populationSean G. Byarsa, Douglas Ewbankb, Diddahally R. Govindarajuc, and Stephen C. Stearnsa,1

aDepartment of Ecology and Evolutionary Biology, Yale University, New Haven, CT 06520-8102; bPopulation Studies Center, University of Pennsylvania,Philadelphia, PA 19104-6299; and cDepartment of Neurology, Boston University School of Medicine, Boston, MA 02118-2526

Edited by Peter T. Ellison, Harvard University, Cambridge, MA, and approved September 16, 2009 (received for review June 25, 2009)

Our aims were to demonstrate that natural selection is operatingon contemporary humans, predict future evolutionary change forspecific traits with medical significance, and show that for sometraits we can make short-term predictions about our future evolu-tion. To do so, we measured the strength of selection, estimatedgenetic variation and covariation, and predicted the response toselection for women in the Framingham Heart Study, a project ofthe National Heart, Lung, and Blood Institute and Boston Univer-sity that began in 1948. We found that natural selection is actingto cause slow, gradual evolutionary change. The descendants ofthese women are predicted to be on average slightly shorter andstouter, to have lower total cholesterol levels and systolic bloodpressure, to have their first child earlier, and to reach menopauselater than they would in the absence of evolution. Selection istending to lengthen the reproductive period at both ends. Tobetter understand and predict such changes, the design of plannedlarge, long-term, multicohort studies should include input fromevolutionary biologists.

evolutionary rates | heritability | Homo sapiens | medical traits

Are contemporary humans experiencing natural selectionand evolving in response to it? The answer to that question

depends on whom one asks. A long tradition in the medicalcommunity (1) holds that natural selection does not operate oncontemporary human populations because medicine keeps“alive many who otherwise would have perished” (2). Noevolutionary biologist would now agree with that claim, fornatural selection works through differential reproductive successrather than simple differential survival, and individuals in con-temporary human populations vary in lifetime reproductivesuccess (LRS). Selection operates on any trait that varies and iscorrelated with LRS, and traits respond to selection with changeacross generations if they vary genetically. But what traits isselection operating on? Do they include the traits treated byphysicians? Previous work (e.g., ref. 3) has shown that human lifehistory traits, most significantly age at first reproduction, arecurrently under selection, but evidence for selection operatingon traits of medical importance is scarce. Here, we reportestimates of natural selection, and the potential genetic responseto selection, in the women of the first two generations of theFramingham Heart Study (FHS) population. The traits weanalyzed include traits of medical significance: total cholesterol(TC), systolic blood pressure (SBP), diastolic blood pressure(DBP), and blood glucose (GLU). We had three general aims:first, to correct the still widespread misconception that naturalselection is not operating on contemporary humans; second, tomake quantitative predictions about future evolutionary changefor specific traits with medical significance; and third, to registerfirmly a point of general cultural interest that follows directlyfrom our first two aims: We are still evolving, and for some traitswe can make short-term predictions about our future evolution.

The Framingham Heart StudyThe FHS was established in 1948 in Framingham, MA, by theNational Heart, Lung, and Blood Institute and Boston Univer-

sity to identify factors that contribute to cardiovascular disease.It is the longest running multigenerational study in medicalhistory. The people originally enrolled in the study were of pre-dominantly European ancestry (20% United Kingdom, 40%Ireland, 10% Italy, 10% Quebec). The original cohort (n =5,209) has been examined every 2 years, a total of 29 timesbetween 1948 and 2008. The offspring cohort (n = 5,124) hasbeen examined approximately every 4 years, a total of eight timesbetween 1971 and 2008 (4). There is also a third generationcohort (n = 4,095) that is not included in this study because manyin it have not yet completed reproduction. At each examinationmany physical and blood chemistry traits are measured and aquestionnaire is administered, yielding data on >70 traits. Dataare deidentified by the FHS and delivered to the NationalInstitutes of Health dbGaP database, from which we down-loaded them for analysis. In this study, we use only the data onindividuals who were measured three or more times.

Measuring Selection in a Multicohort Medical StudyNatural selection has been measured many times in naturalpopulations of animals and plants (5) using methods inspired byRobertson (6), developed by Lande and Arnold (7), and refinedby Janzen and Stern (8), Hereford et al. (9), and others. To applythose methods to contemporary human populations requiresconsideration of several special features of data on humans.Some, such as cultural variation related to education, smoking,and medication, we dealt with as covariates. Others could inprinciple be measured on natural populations of animals andplants but in practice often are not; these include repeatedmeasures on individuals that establish the developmental tra-jectories of multiple traits with age and long-term observationsof populations across several generations that reflect seculardemographic trends. Both make the measurement of traits morecomplex: at what age and in what portion of a secular trend—achange in conditions across time rather than age—should theexpression of the trait be measured? The solution we chose wasto calculate the response surface of each trait for age and timeand to express the measurement of that trait for each individualas an average deviation from that surface (e.g., Fig. 1). Thus, forseveral traits we asked whether through their adult years indi-viduals tended to have higher or lower values than otherindividuals of the same age measured in the same year. Becausemany individuals have been measured repeatedly in the FHS, theresponse surface can be estimated accurately. And how should

This paper results from theArthurM.Sackler Colloquiumof theNationalAcademyofSciences,“Evolution inHealth andMedicine”held April 2–3, 2009, at the National Academyof SciencesinWashington, DC. The completeprogramandaudiofiles ofmost presentations are availableon the NAS web site at www.nasonline.org/Sackler_Evolution_Health_Medicine.

Author contributions: D.R.G. and S.C.S. designed research; S.G.B. performed research;S.G.B. and D.E. analyzed data; and D.R.G. and S.C.S. wrote the paper.


This article is a PNAS Direct Submission.1To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/cgi/content/full/0906199106/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.0906199106 PNAS | January 26, 2010 | vol. 107 | suppl. 1 | 1787–1792

Natural selection in a contemporaryhuman populationSean G. Byarsa, Douglas Ewbankb, Diddahally R. Govindarajuc, and Stephen C. Stearnsa,1

aDepartment of Ecology and Evolutionary Biology, Yale University, New Haven, CT 06520-8102; bPopulation Studies Center, University of Pennsylvania,Philadelphia, PA 19104-6299; and cDepartment of Neurology, Boston University School of Medicine, Boston, MA 02118-2526

Edited by Peter T. Ellison, Harvard University, Cambridge, MA, and approved September 16, 2009 (received for review June 25, 2009)

Our aims were to demonstrate that natural selection is operatingon contemporary humans, predict future evolutionary change forspecific traits with medical significance, and show that for sometraits we can make short-term predictions about our future evolu-tion. To do so, we measured the strength of selection, estimatedgenetic variation and covariation, and predicted the response toselection for women in the Framingham Heart Study, a project ofthe National Heart, Lung, and Blood Institute and Boston Univer-sity that began in 1948. We found that natural selection is actingto cause slow, gradual evolutionary change. The descendants ofthese women are predicted to be on average slightly shorter andstouter, to have lower total cholesterol levels and systolic bloodpressure, to have their first child earlier, and to reach menopauselater than they would in the absence of evolution. Selection istending to lengthen the reproductive period at both ends. Tobetter understand and predict such changes, the design of plannedlarge, long-term, multicohort studies should include input fromevolutionary biologists.

evolutionary rates | heritability | Homo sapiens | medical traits

Are contemporary humans experiencing natural selectionand evolving in response to it? The answer to that question

depends on whom one asks. A long tradition in the medicalcommunity (1) holds that natural selection does not operate oncontemporary human populations because medicine keeps“alive many who otherwise would have perished” (2). Noevolutionary biologist would now agree with that claim, fornatural selection works through differential reproductive successrather than simple differential survival, and individuals in con-temporary human populations vary in lifetime reproductivesuccess (LRS). Selection operates on any trait that varies and iscorrelated with LRS, and traits respond to selection with changeacross generations if they vary genetically. But what traits isselection operating on? Do they include the traits treated byphysicians? Previous work (e.g., ref. 3) has shown that human lifehistory traits, most significantly age at first reproduction, arecurrently under selection, but evidence for selection operatingon traits of medical importance is scarce. Here, we reportestimates of natural selection, and the potential genetic responseto selection, in the women of the first two generations of theFramingham Heart Study (FHS) population. The traits weanalyzed include traits of medical significance: total cholesterol(TC), systolic blood pressure (SBP), diastolic blood pressure(DBP), and blood glucose (GLU). We had three general aims:first, to correct the still widespread misconception that naturalselection is not operating on contemporary humans; second, tomake quantitative predictions about future evolutionary changefor specific traits with medical significance; and third, to registerfirmly a point of general cultural interest that follows directlyfrom our first two aims: We are still evolving, and for some traitswe can make short-term predictions about our future evolution.

The Framingham Heart StudyThe FHS was established in 1948 in Framingham, MA, by theNational Heart, Lung, and Blood Institute and Boston Univer-

sity to identify factors that contribute to cardiovascular disease.It is the longest running multigenerational study in medicalhistory. The people originally enrolled in the study were of pre-dominantly European ancestry (20% United Kingdom, 40%Ireland, 10% Italy, 10% Quebec). The original cohort (n =5,209) has been examined every 2 years, a total of 29 timesbetween 1948 and 2008. The offspring cohort (n = 5,124) hasbeen examined approximately every 4 years, a total of eight timesbetween 1971 and 2008 (4). There is also a third generationcohort (n = 4,095) that is not included in this study because manyin it have not yet completed reproduction. At each examinationmany physical and blood chemistry traits are measured and aquestionnaire is administered, yielding data on >70 traits. Dataare deidentified by the FHS and delivered to the NationalInstitutes of Health dbGaP database, from which we down-loaded them for analysis. In this study, we use only the data onindividuals who were measured three or more times.

Measuring Selection in a Multicohort Medical StudyNatural selection has been measured many times in naturalpopulations of animals and plants (5) using methods inspired byRobertson (6), developed by Lande and Arnold (7), and refinedby Janzen and Stern (8), Hereford et al. (9), and others. To applythose methods to contemporary human populations requiresconsideration of several special features of data on humans.Some, such as cultural variation related to education, smoking,and medication, we dealt with as covariates. Others could inprinciple be measured on natural populations of animals andplants but in practice often are not; these include repeatedmeasures on individuals that establish the developmental tra-jectories of multiple traits with age and long-term observationsof populations across several generations that reflect seculardemographic trends. Both make the measurement of traits morecomplex: at what age and in what portion of a secular trend—achange in conditions across time rather than age—should theexpression of the trait be measured? The solution we chose wasto calculate the response surface of each trait for age and timeand to express the measurement of that trait for each individualas an average deviation from that surface (e.g., Fig. 1). Thus, forseveral traits we asked whether through their adult years indi-viduals tended to have higher or lower values than otherindividuals of the same age measured in the same year. Becausemany individuals have been measured repeatedly in the FHS, theresponse surface can be estimated accurately. And how should

This paper results from theArthurM.Sackler Colloquiumof theNationalAcademyofSciences,“Evolution inHealth andMedicine”held April 2–3, 2009, at the National Academyof SciencesinWashington, DC. The completeprogramandaudiofiles ofmost presentations are availableon the NAS web site at www.nasonline.org/Sackler_Evolution_Health_Medicine.

Author contributions: D.R.G. and S.C.S. designed research; S.G.B. performed research;S.G.B. and D.E. analyzed data; and D.R.G. and S.C.S. wrote the paper.


This article is a PNAS Direct Submission.1To whom correspondence should be addressed. E-mail: [email protected].

This article contains supporting information online at www.pnas.org/cgi/content/full/0906199106/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.0906199106 PNAS | January 26, 2010 | vol. 107 | suppl. 1 | 1787–1792

5000 people & their kids; 70 traits measured every 2-4 years since 1948.


Accounts of human evolution frequently assume that the selective events that shaped us were changes in the external environment, stemming from events beyond human control. For instance, theories of the inception of Homo species emphasize a global trend towards cooler, drier climates, which pushed an arboreal ape out of con-tracting forests into savannah1. Likewise, heat stress in the open is a plausible hypothesis for the evolution of bipedality, hairless skin and sweating2–4. By contrast, lit-tle consideration has been given to the possibility that cultural practices might have transformed the selection pressures acting on humans. This may be because it has only recently been shown that natural selection can bring about substantial changes in genomes that are detectable over thousands of years, as revealed by measurements of the typical rates of response to natural selection among animals in the wild5 and statistical analyses that detected recent rapid adaptation in the human genome6–11.

This traditional conception of human evolution is now being challenged by recent anthropological studies that show that human cultural practices have modified environmental conditions, triggering changes in allele frequencies12,13. In addition, analyses of data from the human genome have revealed numerous genes that have experienced recent positive selection, many of which exhibit functions that imply that they are responses to human cultural practices6–11,14. For instance, several lines of evidence show that dairy farming created the selec-tive environment that favoured the spread of alleles for

adult lactose tolerance12,13,15,16. Estimates for the number of human genes that have been subject to recent rapid evolution range from a few hundred to two thousand: Williamson et al.14 conclude that up to 10% of the human genome may be affected by linkage to targets of posi-tive selection. Although in the majority of cases it is not known what phenotype was the target of the inferred selection, nor which environmental conditions favoured such phenotypes, human cultural practices remain strong candidates, and geneticists are increasingly considering culture as a source of selection on humans17,18.

Such data are consistent with two branches of mathematical evolutionary analysis: gene–culture co- evolutionary theory, which explores how genetic and cultural processes interact over evolutionary time19–23, and niche-construction theory24–30, which investigates the evolutionary impact of the modification of environ-ments by organisms. The models provide hypotheses for, or novel insights into, the evolution of learning, culture, language, intelligence, cooperation, sex differences and mating systems. Analyses of these models have con-firmed that genes and culture could plausibly co-evolve, often revealing patterns and rates of change that are uncharacteristic of more traditional population genetic theory22,31–34. Gene–culture dynamics are typically faster, stronger and operate over a broader range of conditions than conventional evolutionary dynamics, leading some practitioners to argue that gene–culture co-evolution could be the dominant mode of human evolution32–34.

*School of Biology, University of St Andrews, Bute Building, Westburn Lane, St Andrews, Fife KY16 9TS, UK.‡School of Anthropology, University of Oxford, 51/53 Banbury Road, Oxford OX2 6PE, UK.§Institute for Genomic Diversity, Cornell University, Ithaca, New York 14853-2703, USA.||Biology Department, Acadia University, Wolfville, Nova Scotia B4P 2R6, Canada.Correspondence to K.N.L. e-mail: [email protected]:10.1038/nrg2734

How culture shaped the human genome: bringing genetics and the human sciences togetherKevin N. Laland*, John Odling-Smee‡ and Sean Myles§ ||

Abstract | Researchers from diverse backgrounds are converging on the view that human evolution has been shaped by gene–culture interactions. Theoretical biologists have used population genetic models to demonstrate that cultural processes can have a profound effect on human evolution, and anthropologists are investigating cultural practices that modify current selection. These findings are supported by recent analyses of human genetic variation, which reveal that hundreds of genes have been subject to recent positive selection, often in response to human activities. Here, we collate these data, highlighting the considerable potential for cross-disciplinary exchange to provide novel insights into how culture has shaped the human genome.

REVIEWS

NATURE REVIEWS | GENETICS VOLUME 11 | FEBRUARY 2010 | 137

© 20 Macmillan Publishers Limited. All rights reserved10








REVIEWS










REVIEWS




Genetics of our behavior?


Evolutionary Psychology

• Generosity higher if affects reputation

• Pheromones help identify our mates



Summary

• Human evolution is complicated but fascinating!

• (just like any other species!!)


For more info• http://humanorigins.si.edu/ (Smithonian Institution)

• PBS Nova Becoming Human (on youtube)

• Stoneking & Krause. Learning about human population history from ancient and modern genomes. Nature Reviews Genetics 2011.


http://humanorigins.si.edu/

http://humanorigins.si.edu/

Any questionsAbout first lectures?


http://qmplus.qmul.ac.uk/course/view.php?id=3972

Week 1: YW – Introduction, Historical context, Neo DarwinismWeek 2: YW – Geological Aspects, Drivers of Evolution, Levels of EvolutionWeek 3: YW – Fossils, DNA and MoleculesWeek 4: YW – Human Evolution Week 5: DH – Evolution of Sex, Sexual SelectionWeek 6: AH – Genetic Basis of Evolution Week 7: Mid semester break, no lectures.Week 8: AH – Founder Effects, Genetic Drift + Computer Practical (tues or thurs afternoon) Week 9: AH – Mutation, Selection and Gene SelectionWeek 10: DH – Systematics, SpeciationWeek 11: DH – Evolution of Parasites, AntibioticsWeek 12: DH – Convergence, Revision Session

Semester A: Evolution

20% WorkshopFinal Grade:

80% Exam

In Jean Smith’s Timetable; not yet on “SMART”




Date post:	28-Jan-2015
Category:	Education
Upload:	yannick-wurm
View:	137 times
Download:	25 times

Evolution - Week 4: Human evolution

Education