Whole-Genome Sequencing o Na i e Sheep P o ides Insigh s
in o Rapid Adap a ions o Ex eme En i onmen s
Ji Yang,
†,1
Wen-Rong Li,
†,2
Feng-Hua L ,
†,1
San-Gang He,
†,2
Shi-Lin Tian,
†,3
Wei-Feng Peng,
1,4
Ya-Wei Sun,
2,5
Yong-Xin Zhao,
1,4
Xiao-Long Tu,
3
Min Zhang,
1,6
Xing-Long Xie,
1,4
Yu-Tao Wang,
7
Jin-Quan Li,
8
Yong-Gang Liu,
9
Zhi-Qiang Shen,
10
Feng Wang,
11
Guang-Jian Liu,
3
Hong-Feng Lu,
3
Juha Kan anen,
12,13
Jian-Lin Han,
14,15
Meng-Hua Li,*
,1
and Ming-Jun Liu*
,2
1
CAS Key Labo a o y o Animal Ecology and Conse a ion Biology, Ins i u e o Zoology, Chinese Academy o Sciences (CAS), Beijing,
China
2
Animal Bio echnology Resea ch Ins i u e, Xinjiang Academy o Animal Science, U umqi, China
3
No ogene Bioin o ma ics Ins i u e, Beijing, China
4
Uni e si y o Chinese Academy o Sciences (UCAS), Beijing, China
5
College o Animal Science and Technology, Shihezi Uni e si y, Shihezi, China
6
School o Li e Sciences, Uni e si y o Science and Technology o China, He ei, China
7
College o Biological and Geog aphic Sciences, Kashga Uni e si y, Kashga , China
8
College o Animal Science, Inne Mongolia Ag icul u al Uni e si y, Hohho , China
9
College o Animal Science and Technology, Yunnan Ag icul u al Uni e si y, Kunming, China
10
Shandong Binzhou Academy o Animal Science and Ve e ina y Medicine, Binzhou, China
11
Ins i u e o Sheep and Goa Science, Nanjing Ag icul u al Uni e si y, Nanjing, China
12
G een Technology, Na u al Resou ces Ins i u e Finland (Luke), Jokioinen, Finland
l3
Depa men o En i onmen al and Biological Sciences, Uni e si y o Eas e n Finland, Kuopio, Finland
14
CAAS-ILRI Join Labo a o y on Li es ock and Fo age Gene ic Resou ces, Ins i u e o Animal Science, Chinese Academy o Ag icul u al
Sciences (CAAS), Beijing, China
15
In e na ional Li es ock Resea ch Ins i u e (ILRI), Nai obi, Kenya
†
These au ho s con ibu ed equally o his wo k.
*Co esponding au ho : E-mail: [email p o ec ed]n; [email p o ec ed].
Associa e edi o : Yuseob Kim
Abs ac
Global clima e change has a signi ican e ec on ex eme en i onmen s and a p o ound in luence on species su i al.
Howe e , li le is known o he genome-wide pa e n o li es ock adap a ions o ex eme en i onmen s o e a sho ime
ame ollowing domes ica ion. Sheep (O is a ies) ha e become well adap ed o a di e se ange o ag oecological zones,
including ce ain ex eme en i onmen s (e.g., pla eaus and dese s), du ing hei pos -domes ica ion (app oxima ely 8–9
kya) mig a ion and di e en ia ion. He e, we gene a ed whole-genome sequences om 77 na i e sheep, wi h an a e age
e ec i e sequencing dep h o 5 o 75 samples and 42 o 2 samples. Compa a i e genomic analyses among sheep
in con as ing en i onmen s, ha is, pla eau (>4,000 m abo e sea le el) e sus lowland (<100 m), high-al i ude egion
(>1500 m) e sus low-al i ude egion (<1300 m), dese (<10 mm a e age annual p ecipi a ion) e sus highly humid
egion (>600 mm), and a id zone (<400 mm) e sus humid zone (>400 mm), de ec ed a no el se o candida e genes as
well as pa hways and GO ca ego ies ha a e pu a i ely associa ed wi h hypoxia esponses a high al i udes and wa e
eabso p ion in a id en i onmen s. In addi ion, candida e genes and GO e ms unc ionally ela ed o ene gy me abolism
and body size a ia ions we e iden i ied. This s udy o e s no el insigh s in o apid genomic adap a ions o ex eme
en i onmen s in sheep and o he animals, and p o ides a aluable esou ce o u u e esea ch on li es ock b eeding in
esponse o clima e change.
Key wo ds: ex eme en i onmen , apid adap a ion, O is a ies,whole-genome sequencing,clima e change.
In oduc ion
Globalclima echangehasaconside ableimpac ono gan-
isms, including li es ock (B own and Funk 2008;Ho mann
2010). In pa icula , he in luence o clima e change on li ing
o ganisms is exace ba ed a he in e ace o ex eme en i on-
men s, such as on pla eaus and in dese egions (Eas e ling
e al. 2000). Wi hin his con ex , i is impo an o unde s and
he gene ic basis o well-adap ed local li es ock b eeds in
ex eme en i onmen s o de elop app op ia e b eeding
A icle
ßThe Au ho 2016. Published by Ox o d Uni e si y P ess on behal o he Socie y o Molecula Biology and E olu ion.
This is an Open Access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion Non-Comme cial License
(h p://c ea i ecommons.o g/licenses/by-nc/4.0/), which pe mi s non-comme cial e-use, dis ibu ion, and ep oduc ion in any
medium, p o ided he o iginal wo k is p ope ly ci ed. Fo comme cial e-use, please con ac [email p o ec ed] Open Access
2576 Mol. Biol. E ol. 33(10):2576–2592 doi:10.1093/molbe /msw129 Ad ance Access publica ion July 8, 2016
p og ams unde scena ios o u u e clima e change (Howden
e al. 2007;Lobell e al. 2008). A e domes ica ion in he
Fe ile C escen 8,000–9,000 yea s ago (8–9 kya; Legge
1996), sheep (O is a ies) sp ead and became adap ed o a
wide ange o ag oecological condi ions, especially hose dis-
ibu ed on pla eaus o in dese egions, which a e sensi i e
oclima echange(Eas e ling e al. 2000;L e al. 2014). Thus,
hese animals p o ide an excellen model o gain no el in-
sigh s in o gene ic mechanisms unde lying he apid adap a-
ions o li es ock o ex eme en i onmen s wi hin a sho
pe iod o ime.
In ecen yea s, o cha ac e ize adap i e gene ic a ia ions,
whole-genome sequencing s udies ha e been pe o med on a
wide ange o o ganisms ha li e in ha sh o ex eme en i-
onmen s (Simonson e al. 2010;Mackelp ang e al. 2011;
Schein eld e al. 2012;Cho e al. 2013;Ge e al. 2013;Ma
e al. 2013;Qu e al. 2013;Liu e al. 2014). S udies conduc ed
on li es ock a e limi ed, al hough hey include wo k on
adap a ions o high al i udes in yak (Bos g unniens;Qiu
e al. 2012), Tibe an mas i (Canis lupus amilia is;Gou
e al. 2014)andTibe anchicken(Wang e al. 2015), ho
and a id en i onmen s in goa and sheep (Kim e al. 2016),
se e e dese condi ions in domes ic Bac ian camel (Ji imu u
e al. 2012;Wu e al. 2014) and suba c ic cold en i onmen s in
Yaku ian ho se (Lib ado e al. 2015). Howe e , o ou knowl-
edge, no s udy has cha ac e ized he apid gene ic adap a-
ions o li es ock o a ious ex eme en i onmen s based on
whole-genome sequences.
Ou o he dispe sal cen e in he Mongolian egion,
Chinese na i e sheep b eeds only ha e di e ged o se e al
housands o yea s (e.g., L e al. 2015). P e ious in es iga ions
showed signi ican di e ences in hema ologic pa ame e s
(e.g., e y h ocy e coun , hema oc i , and hemoglobin; Jiang
e al. 1991;Liu and Chen 2006)andbodysize(Du 2011)
be ween Tibe an sheep on he Qinghai–Tibe an Pla eau
and hose a ela i ely low al i udes as well as obus mo -
phology (i.e., good body cons i u ion; Du 2011)insheepo i-
gina ing om he Taklimakan Dese egion.
In his s udy, we sequenced he whole genomes o 77
sheep (O. a ies) including hose om habi a s in ex eme
(o ha sh)en i onmen s:Tibe an a eas on he Qinghai–
Tibe an Pla eau (de ined as ‘pla eau’, al i ude >4,000 m abo e
sea le el), high-al i ude egion (al i ude >1,500 m),
Taklimakan Dese egion (de ined as ‘dese ’, a e age annual
p ecipi a ion <10 mm), and a id zone (<400 mm, ep esen -
ing a id and semi-a id egions; Piao e al. 2010; ig. 1Aand B,
supplemen a y ables S1 and S2,Supplemen a y Ma e ial on-
line). The se o samples ep esen ed 21 na i e b eeds o di -
e en gene ic and geog aphic o igins in China
(supplemen a y able S1,Supplemen a y Ma e ial online).
Th ough compa isons o he genomes o sheep om ex eme
en i onmen s wi h hose om con as ing en i onmen s, we
aimed o iden i y he candida e genes, unc ional Gene
On ology (GO) ca ego ies and signaling pa hways esponsible
o he apid adap a ions (i.e., o e housands o yea s) o
sheep o pla eau and dese en i onmen s. In addi ion, o
elucida e he e olu iona y his o y o Chinese na i e sheep,
a comp ehensi e analysis o he genomic di e si y, popula ion
s uc u e and demog aphic his o y o hese animals was pe -
o med based on genomic da a.
Resul s and Discussion
Genomic Va ian s
We gene a ed he genomic sequences o 77 na i e sheep and,
o genomic compa ison and as phylogene ic ou g oups,
h ee wild species o he sub amily Cap inae, izO. a.
musimon,O is ammon polii and Cap a ibex (one animal
om each species). This yielded 1,150 Gb o aligned high-
quali y genomic da a a a o al o 438e ec i e sequence
dep h o he subsequen analyses.
An a e age o 94.66% (92.14–95.30%) o he SNPs iden i-
ied in he 77 na i e sheep we e alida ed in he sheep dbSNP
da abase (supplemen a y able S3,Supplemen a y Ma e ial
online), indica ing he high eliabili y o he called SNP a i-
a ions in his s udy (e.g., Pa k e al. 2015). Fo he 3,720,550
SNPs no con i med by dbSNP da abase, 1,582,868 SNPs we e
u he alida ed by hei p esence in 2–77 indi iduals o he
samples. O he common SNPs o ou sequencing da a and
he O ine 50K Beadchip a ay in he 33 na i e sheep (7,200–
15,016 SNPs in each indi idual), an a e age 95.51% (89.46–96.
21%) o he SNP geno ypes we e consis en (supplemen a y
able S4 and ig. S1, Supplemen a y Ma e ial online), he eby
demons a ing he high quali y o ou SNP calling. The 4.49%
disc epancy was esul ed om geno yping e o s and missing
da a in he whole genome sequencing p ocess (supplemen a y
able S5,Supplemen a y Ma e ial online). Mo eo e , a
la ge numbe o sha ed SNPs (15.58–17.48 million) be-
ween di e en sheep g oups om di e en egions (i.e.,
Qinghai–Tibe an Pla eau, Yunnan–Kweichow Pla eau,
No he n and Eas e n China; supplemen a y able S6,
Supplemen a y Ma e ial online) also suppo o he a-
lidi y o he SNP calling.
We ob ained 21.26 million high-quali y SNPs o 77 na i e
sheep, 2.77 million o O.a.musimon, 4.71 million o O.a.
polii, and 10.10 million o C.ibex (supplemen a y able S7,
Supplemen a y Ma e ial online). The e we e 15.81 million
high-quali y SNPs unique o he na i e sheep (supplemen a y
able S7,Supplemen a y Ma e ial online). Based on one indi-
idual pe species, he o e all numbe s o SNPs pe bp in he
exonic, in onic, in e genic egions and ch omosomes o do-
mes ic sheep (a e aged o e 77 indi iduals) we e highe han
hose o O.a.musimon, bu lowe han hose o O.a.polii and
C.ibex (supplemen a y able S8,Supplemen a y Ma e ial on-
line). O he h ee iden i ied gene ic g oups o na i e sheep,
No he n and Eas e n Chinese b eeds possessed he la ges
amoun s o he o al (19,227,904) and unique (784,796) SNPs
(supplemen a y able S6 and ig. S2A, Supplemen a y Ma e ial
online), indica ing hei o e all ela i ely high genomic di-
e si y and ea ly popula ion his o y (L e al. 2015). Mos o
he high-quali y SNPs in he indi idual genome o sheep
we e loca ed in in e genic egions (2.678 million, 71.83%),
wi h only 0.76% (0.028 million) loca ed in exonic egions
(supplemen a y able S8,Supplemen a y Ma e ial online).
A o al o 11,167 nonsynonymous SNPs and 16,879 synon-
ymous SNPs we e loca ed wi hin exons, which esul ed in a
Rapid Adap a ion o Sheep o Ex eme En i onmen s .doi:10.1093/molbe /msw129 MBE
2577
AB
CD
E
G
F
FIG.1.Geog aphic dis ibu ion and popula ion gene ics analyses o 21 na i e sheep b eeds. (A) Sampling si es in his s udy. A o al o 77 Chinese
sheep ep esen ing 21 na i e b eeds we e included. The ele a ion (m) o he s udy a ea is also isualized. (B) Geog aphic a ia ion o he annual
mean p ecipi a ion (mm). P ecipi a ion da a du ing he 21 yea s om 1981 o 2001 we e e ie ed om he Abdus Salam In e na ional Cen e o
Theo e ical Physics, I aly (ICTP, h ps://www.ic p.i , las accessed Janua y 5, 2015). The 200, 400, and 800 mm a e age annual p ecipi a ion lines a e
also isualized. (C) Diag am o he a e age sha ed SNPs be ween O.a ies indi iduals as well as be ween O.a ies indi iduals and he indi iduals o he
h ee wild species (O.a.musimon,O.a.polii, and C.ibex). (D) Decay o linkage disequilib ium (LD) in he Chinese na i e sheep b eeds, wi h one line
pe b eed. (E) NJ ee cons uc ed using p-dis ances be ween indi iduals. The h ee wild species a e used as ou g oups. (F) P incipal componen s 1
and 2 o he 77 na i e sheep. (G) Popula ion gene ic s uc u e o he 77 na i e sheep in e ed om he p og am FRAPPE 1.1. The leng h o each
colo ed segmen ep esen s he p opo ion o he indi idual genome in e ed om ances al popula ions (K¼2–4). See supplemen a y able S1,
Supplemen a y Ma e ial online o he abb e ia ions o he b eeds and indi iduals.
Yang e al. .doi:10.1093/molbe /msw129 MBE
2578
nonsynonymous/synonymous a io o 0.662. This a io was
wi hin he ange o a ios (0.40–1.42) o humans and o he
animals (e.g., mou lon, a gali, ibex, and goa ; supplemen a y
able S9,Supplemen a y Ma e ial online).
The e we e 15.98% (10.40–27.56%) o SNPs sha ed
be ween indi iduals wi hin b eeds, 14.20% (8.86–26.41%) be-
ween b eeds, 11.08% (9.13–13.05%) o O.a ies–O.a.musi-
mon, 7.87% (6.28–10.13%) o O.a ies–O.a.polii,and1.60%
(1.32–2.07%) o O.a ies–O.a.musimon–O.a.polii ( ig. 1C).
Fo he h ee wild species, he sha ed p opo ions o SNPs
we e 6.93%, 2.65%, 4.28%, and 0.64% o O.a.musimon–O.a.
polii,O.a.musimon–C.ibex,O.a.polii–C.ibex and O.a.
musimon–O.a.polii-C.ibex, espec i ely. These s a is ics
showed he p opo ions o SNPs wi h a gi en axonomic
ange (i.e., O ine species, O.a ies and b eeds) and hus o
di e en o igins (i.e., om p eda ing he di e gence o o ine
species o p edomes ic, and hen o pos da ing b eed di e -
gence). Mo eo e , O.a ies–O.a ies showed he la ges a e age
numbe o sha ed SNPs pe pai o indi iduals, ollowed by
O.a ies–O.a.musimon,O.a ies–O.a.polii,andO.a ies–C.ibex
(supplemen a y able S7,Supplemen a y Ma e ial online).
The a e age numbe s o p i a e SNPs pe pai o indi iduals
anged om 2.11 million o O.a ies–O.a.musimon o 9.77
million o O.a.musimon–C.ibex (supplemen a y able S10,
Supplemen a y Ma e ial online).
In addi ion, we iden i ied an a e age o 264,871 indels
(<100 bp) pe indi idual o he na i e sheep, 192,315 o
O.a.musimon, 354,619 o O.a.polii and 755,867 o C.ibex
(supplemen a y able S11,Supplemen a y Ma e ial online).
Fo he wo high-co e age samples (ZNQ24 and LOP41),
we de ec ed 177,371 copy numbe a ia ions (CNVs,
200 bp–5 Mb; F eeman e al. 2006)(supplemen a y able
S12,Supplemen a y Ma e ial online) and 156,134 s uc u al
a ia ions (SVs, 100 bp–ch omosome le el; Alkan e al. 2011)
which included inse ions, dele ions, in e sions, in a-
ch omosomal ansloca ions and in e -ch omosomal anslo-
ca ions (supplemen a y able S13,Supplemen a y Ma e ial
online). Mos o he indels, CNVs and SVs we e loca ed in
in e genic egions.
Pa e ns o Genomic Va ia ion and Linkage
Disequilib ium
The genome-wide a e age h
p
alue o hena i esheepwas
2.28 10
3
(1.9–2.5 10
3
), which was wi hin he anges o
he pa ame e s o o he animals and humans (supplemen a y
able S14,Supplemen a y Ma e ial online). The genomic a -
ia ion pa ame e s [a e age h
p
alue, a e age numbe o e-
gions o homozygosi y (ROHs) and mean ROH size] o he
h ee sheep g oups showed cong uen pa e ns. In o he
wo ds, he No he n and Eas e n Chinese b eeds showed
he g ea es genomic di e si y (h
p
) and exhibi ed he ewes
ROH numbe (879) and he smalles a e age ROH size (171.
34 kb), whe eas he lowes genomic a ia ions we e ob-
se ed in he b eeds om he Yunnan–Kweichow Pla eau
(supplemen a y ables S15 and S16 and ig. S2B–G,
Supplemen a y Ma e ial online). Inb ed indi iduals we e
no obse ed in he 77 na i e sheep samples acco ding o
he IBS sco e (IBS <0.9). The b eeds om he Qinghai–
Tibe an Pla eau (e.g., ZLZ and ZCD) and he Yunnan–
Kweichow Pla eau (e.g., WNS and SPS) showed an o e all
slow decay a e and a high le el o LD, whe eas he b eeds
om No he n and Eas e n China exhibi ed a apid decay
a e and a low le el o LD ( ig. 1D). The di e ences in he
genome-wide LD pa e ns be ween b eeds likely e lec ed
la ge e ec i e popula ion sizes (N
e
) o he No he n and
Eas e n Chinese b eeds compa ed wi h he o he b eeds.
O e all, he pa e ns o genomic di e si y and LD s ongly
sugges ed a no he n o igin o Chinese sheep, om which
Qinghai–Tibe an and Yunnan–Kweichow b eeds we e sub-
sequen ly de i ed. This esul is consis en wi h he mig a-
ion ou es o eas e n Eu asian sheep, being om no he n
China o sou he n China, which we e deduced in ou ecen
mi ogenomic s udy (L e al. 2015).
Popula ion Gene ic S uc u e
The neighbo -joining (NJ) ee e ealed s ong clus e ing o
he na i e sheep in o h ee gene ic g oups ha p esen ed a
low le el o gene ic di e en ia ion (mean F
ST
¼0.014–0.040;
supplemen a y able S17,Supplemen a y Ma e ial online):
Qinghai–Tibe an b eeds (n¼29,ZCD,ZNQ,ZRK,ZLZ,
GDS, GZS, and MXS), Yunnan–Kweichow b eeds (n¼20,
TCS, WNS, SPS, and DQS) and No he n and Eas e n
Chinese b eeds (n¼28, BRK, LOP, NMS, HUS, WDS, ALS,
BYK, KAZ, HTS, and TSK; ig. 1E,supplemen a y ig. S3,
Supplemen a y Ma e ial online). A p incipal componen
analysis (PCA) and a Bayesian model-based clus e ing analysis
p o ided addi ional co obo a ing e idence o hese
g oupings ( ig. 1Fand G,supplemen a y igs. S4 and S5,
Supplemen a y Ma e ial online). In he clus e ing analysis,
when K¼2, he na i e sheep we e gene ically di ided
in o he Yunnan–Kweichow b eeds and he emaining
b eeds; when K¼4, he No he n and Eas e n Chinese
b eeds, he Qinghai–Tibe an b eeds and popula ion
ZLZ o he Tibe an sheep o med h ee addi ional gene ic
clus e s ( ig. 1G). We obse ed a clea signa u e o gene ic
admix u e be ween he sheep b eeds om he Qinghai–
Tibe an Pla eau and No he n and Eas e n China as well
as gene ic in og ession om No he n and Eas e n
Chinese b eeds in o he Yunnan–Kweichow b eeds ( ig.
1G,supplemen a y ig. S5,Supplemen a y Ma e ial on-
line). Compa ed wi h he adi ional iew ha Chinese
na i e sheep we e de i ed om h ee old b eeds (i.e.,
Kazakh, Mongolian and Tibe an sheep; Du 2011), ou esul s
p o ided e idence ha b eeds om he Qinghai–Tibe an
Pla eau, Yunnan–Kweichow Pla eau, and No he n and
Eas e n China a e gene ically dis inc ( ig. 1G). This inding
is consis en wi h a ecen s udy on 10 ep esen a i e
Chinese sheep b eeds geno yped using he 50K SNP
Beadchip (Wei e al. 2015).
Demog aphic His o y and Popula ion Admix u e
The wo high-co e age samples (ZNQ24 and LOP41)
exhibi ed conco dan demog aphic ajec o ies, wi h wo ap-
pa en expansions and wo se e e bo lenecks ha mi o ed
he glacial cycles and sea le el luc ua ions ha a ec ed he
a eas ( ig. 2A). The i s bo leneck occu ed 0.6 Mya
Rapid Adap a ion o Sheep o Ex eme En i onmen s .doi:10.1093/molbe /msw129 MBE
2579
(million yea s ago; ig. 2A) and is consis en wi h he
Naynayxungla glacia ion (0.78–0.50 Mya), which was he
mos ex ensi e glacia ion du ing he Qua e na y Pe iod
(Zheng e al. 2002;Lehmkuhl and Owen 2005;Ehle s and
Gibba d 2007). A ha ime, he sea le el equen ly luc u-
a ed and was main ained a an o e all low le el ( ig. 2A). A e
he e ea o he Naynayxungla glacia ion, he ances al
sheep popula ion expanded and eached a pinnacle
(0.18 Mya; ig. 2A) du ing he Penul ima e glacia ion
(0.30–0.13 Mya; Zheng e al. 2002;Lehmkuhl and Owen
2005;Ehle s and Gibba d 2007). The cold-clima e in e al
and ising sea le el a his s age could ha e con ibu ed o
a popula ion expansion because an inc ease in g assland was
likely unde such en i onmen al condi ions (Lo enzen e al.
2011). The second bo leneck (70 kya; ig. 2A)wasde ec ed
owa ds he end o he in e glacial pe iod (0.13–0.07 Mya;
Zheng e al. 2002;Lehmkuhl and Owen 2005;Ehle s and
Gibba d 2007), which p esen ed en i onmen al condi ions
simila o ha o he p esen (O lando e al. 2013). The an-
ces al sheep popula ion size hen peaked again a 15 kya
du ing he Las Glacial Maximum (LGM) because he glacia-
ions we e less ex ensi e and he g asslands had expanded
(Zheng e al. 2002;Lehmkuhl and Owen 2005;Ehle s and
Gibba d 2007; ig. 2A). Subsequen ly, he popula ion size g a-
dually dec eased o a small numbe ( ig. 2A), which was mos
likely he esul o clima e wa ming and associa ed g assland
FIG.2.Demog aphic his o y o he sheep popula ion. (A) PSMC analysis esul s o he ep esen a i e indi iduals sequenced a a high ead
co e age (42) exhibi in e ed a ia ions in N
e
o e he las 10
6
yea s. The N
e
o mou lon, a gali, ibex, dog, ho se, and wild boa a e also escaled.
The glacia ion pe iods, a mosphe ic su ace ai empe a u e (C) and global ela i e sea le el da a o e he las 10
6
yea s a e included. (B)@
a
@
i
analysis showing he demog aphic his o y o Chinese na i e sheep om 4,000 yea s ago o he p esen . Two popula ion di e gences occu ed a
3,272 ( 1) and 2,134 ( 2) yea s ago. The a e age numbe o mig an s pe yea be ween g oups is shown be ween he black a ows. (C) Phylogene ic
ne wo k o he in e ed ela ionships among he 21 na i e b eeds wi h h ee in e -g oup mig a ion edges. The colo ed egions in he phylogene ic
ee ep esen h ee in e ed gene ic g oups. A ows indica e mig a ion e en s, and a spec um o hea colo s indica e he mig a ion weigh s o he
mig a ion e en s. The scale ba shows 10 he a e age s anda d e o o he en ies in he sample co a iance ma ix. NEC, No he n and Eas e n
Chinese b eeds; QT, Qinghai–Tibe an b eeds; YK, Yunnan–Kweichow b eeds. Fo he abb e ia ions o he b eeds, see supplemen a y able S1,
Supplemen a y Ma e ial online.
Yang e al. .doi:10.1093/molbe /msw129 MBE
2580

con ac ions a e he LGM (Lo enzen e al. 2011). Despi e
ela i ely low es ima es o N
e
, he pai wise sequen ially
Ma ko ian coalescen (PSMC) esul s o he low-co e age
samples (supplemen a y ig. S6,Supplemen a y Ma e ial on-
line) showed demog aphic ends ha we e consis en wi h
hose in e ed om he wo high-co e age samples ( ig. 2A).
In gene al, ou indings a e consis en wi h p e ious esul s,
indica ing his o ical popula ion expansion e en s o Chinese
sheep as in e ed om m DNA sequence a iabili y (Wang
e al. 2007;Zhao e al. 2013). Compa ed wi h o he animals
(e.g., ho se, O lando e al. 2013; wild boa , Li e al. 2013;dog,
F eedman e al. 2014; mou lon, a gali, and ibex in his s udy),
he demog aphic end o sheep is mos simila o ho se (e.g.,
close imings o peaks and bo lenecks; ig. 2A)andmou lon
(supplemen a y ig. S6,Supplemen a y Ma e ial online).
These simila demog aphic ends migh ha e been a i-
bu ed o he simila habi a equi emen s o hese he bi-
o es. The di e ences in demog aphic shapes such as he
ex en o he changes in e ec i e popula ion sizes and iming
o expansion/con ac ion we e obse ed be ween sheep and
he wo wild Cap inae species (i.e., a gali and ibex; supplemen
a y ig. S6,Supplemen a y Ma e ial online). This may esul
om smalle N
e
and highe suscep ibili y o en i onmen al
changes (e.g., habi a agmen a ion) o he wild he bi o es
compa ed wi h sheep and a ious en i onmen al luc ua ions
in geog aphic egions whe e hey inhabi ed.
Among he i e di usion app oxima ion o demog aphic
in e ence (@a@i) models es ed, model 3 achie ed a maxi-
mum log-likelihood alue o 16643 and was selec ed as
he op imal model (supplemen a y able S18,
Supplemen a y Ma e ial online). This model showed ha
No he n and Eas e n Chinese b eeds di e ged om he e-
maining b eeds 3,200 yea s ago, wi h an es ima ed N
e
o
24,220 o No he n and Eas e n Chinese b eeds and 498 o
he o he b eeds ( ig. 2B,supplemen a y able S19,
Supplemen a y Ma e ial online). App oxima ely 2,100 yea s
ago, b eeds in a eas o he han No he n and Eas e n China
u he di e ged in o he Qinghai–Tibe an b eeds and
Yunnan–Kweichow b eeds, bo h o which ha e g adually in-
c eased and exhibi ed es ima ed N
e
alues om 2,881 o 5,122
and 1,707 o 2,667, espec i ely ( ig. 2B,supplemen a y able
S19,Supplemen a y Ma e ial online). The cu en N
e
alue o
No he n and Eas e n Chinese b eeds is 4,873, which has ob-
iously dec eased since hei ini ial appea ance (24,220; ig. 2B,
supplemen a y able S19,Supplemen a y Ma e ial online).
An o e all low o mode a e le el o gene low (0.082–11.
791 mig an s pe yea ) was de ec ed be ween he sheep
g oups ( ig. 2B,supplemen a y able S19,Supplemen a y
Ma e ial online).
The maximum-likelihood (ML) ee wi hou mig a ion
e en s in e ed om he T eeMix analysis di ided he 77 na-
i e sheep in o h ee clus e s ( ig. 2C), which a e simila o he
popula ion s uc u ing pa e ns iden i ied om he phyloge-
ne ic ee, PCA and gene ic s uc u e analysis ( ig. 1E–G).
When po en ial mig a ion edges (i.e., mig a ion e en s be-
ween he b anches) we e added o he ML ee, s ong mi-
g a ion e en s we e de ec ed among he h ee clus e s. Up o
98.77% o he a iance be ween b eeds was explained by a
model wi h i e mig a ion e en s (supplemen a y ig. S7,
Supplemen a y Ma e ial online). In he model, we obse ed
h ee mig a ion edges among clus e s om No he n and
Eas e n China o he Yunnan–Kweichow Pla eau, om he
Yunnan–Kweichow Pla eau o he Qinghai–Tibe an Pla eau
and om he Qinghai–Tibe an Pla eau o No he n and
Eas e n China ( ig. 2C). This esul is consis en wi h he
in e -g oup gene low ound in he @a@ianalysis and he
gene ic admix u e be ween sheep g oups obse ed in he
gene ic s uc u e analysis.
Genome-Wide Selec i e Sweep Tes
Using he op 5% o F
ST
alues and h
p
a io cu o s
(Z(F
ST
)>1.834, 1.859, 1.731, and 1.823 and log
2
(h
p
a-
io) >0.352, 0.277, 0.247 and 0.198 o he Con ol g oup
(HUS and WDS om Eas e n China; see supplemen a y able
S1,Supplemen a y Ma e ial online)/Tibe an g oup, Con ol
g oup/Taklimakan Dese g oup, Low-al i ude g oup/High-al-
i ude g oup and Humid g oup/A id g oup, espec i ely; igs.
3Aand 4A), we iden i ied 731 candida e genes associa ed
wi h pla eau adap a ions (>4,000 m; supplemen a y able
S20,Supplemen a y Ma e ial online), 452 candida e genes
associa ed wi h dese adap a ions (a e age annual p ecipi a-
ion <10 mm; supplemen a y able S21,Supplemen a y
Ma e ial online), 465 candida e genes in ol ed in high-
al i ude adap a ions (>1,500 m; supplemen a y able S22,
Supplemen a y Ma e ial online) and 603 candida e genes in-
ol ed in a id adap a ions (a e age annual p ecipi a-
ion <400 mm; supplemen a y able S23,Supplemen a y
Ma e ial online) o he na i e sheep. In addi ion, 261 ( op
5% ou lie s, XP-EHH alue >1.076) and 145 genes ( op 5%
ou lie s, XP-EHH alue >0.897) we e posi i ely selec ed in XP-
EHH analysis, and 147 ( op 1% ou lie s, log
10
(P)>0.433)
and 95 genes ( op 1% ou lie s, log
10
(P)>0.456) we e po-
si i ely selec ed in LFMM analysis ega ding he pla eau and
dese en i onmen s, espec i ely. O e all, 112 and 75 genes
we e posi i ely selec ed ac oss all he ou me hods unde he
pla eau and dese en i onmen s, espec i ely ( igs. 5 and 6,
supplemen a y ables S20 and S21,Supplemen a y Ma e ial
online).
The phylogene ic ee econs uc ed based on he SNPs o
he candida e genes o Tibe an sheep showed ha mos o he
77 na i e sheep (71/77, 92.21%) we e co ec ly assigned in o
sepa a ed clades be ween he high-al i ude (i.e., he Qinghai–
Tibe an Pla eau and Yunnan–Kweichow Pla eau) and low-
al i ude egions (supplemen a y ig. S8A,Supplemen a y
Ma e ial online). Simila ly, he phylogene ic ee ha included
he SNPs o he candida e genes o b eeds om he Taklimakan
Dese egion la gely di ided he 77 sheep (74/77, 96.10%) in o
wo clades be ween a id (i.e., a e age annual p ecipi a-
ion <200mm)andnon-a id egions(supplemen a y ig.
S8B,Supplemen a y Ma e ial online). This opological pa e n
was di e en om ha in e ed om he genome-wide SNPs
( ig. 1E), which suppo ed he c edibili y o he candida e
genes unde selec ion. Rega ding he la ge-e ec SNP anal-
yses o he wo ca ego ies o candida e genes, we obse ed
signi ican ly (P<0.05) highe equency di e ences be-
ween he con as ing g oups han be ween he o he
Rapid Adap a ion o Sheep o Ex eme En i onmen s .doi:10.1093/molbe /msw129 MBE
2581
pai wise g oups (supplemen a y ig. S9,Supplemen a y
Ma e ial online). Mo eo e , he Z(F
ST
) alues(supplemen a y
ig. S10,Supplemen a y Ma e ial online) and log
2
(h
p
a io)
alues (supplemen a y ig. S11,Supplemen a y Ma e ial on-
line) we e much highe o he selec ed genomic egions o
he sheep om he pla eau egion, dese egion, high-al i ude
a ea and a id zone han he alues a he whole-genome le el.
Compa ed wi h he o e lap expec ed om chance, he e was
a signi ican excess o o e lapping selec i e signals sha ed be-
ween he candida e genes iden i ied he e and he p ede ined
gene panel (i.e., he p e iously published candida e genes in
o he mammalian species unde simila ex eme
-4 -2 02
20
40
60
80
100
Cumula i e (%)
0 20406080100
Cumula i e (%)
2
4
6
8
10
F equency (%)
0.2 0.4
0510
4
log2(θπ·con ol/θπ·Tibe an)F equency (%)
0
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
-1.0
0.0
3.0
FST
log2(θπ·con ol/θπ·Tibe an)
Tajimas’D
FST
log2(θπ·con ol/θπ·Tibe an)
Tibe an sheep
Con ol sheep
O is a ies-M
O is a ies-W
Bos au us
Tu siops unca us
Pan oglody es
Go illa go illa go illa
Homo sapiens
Calli h ix jacchus
Mic ocebus mu inus
P e opus ampy us
Loxodon a a icana
Ca ia po cellus
O yc olagus cuniculu
s
Mus musculus
0.000.10
CCCTGAGGGAA
CCCTGCGGGAA
CCCTGCGGGAA
CCCTGCGGGAA
CCCTGCGGGAG
CCCTGCGGGAG
CCCTGCGGGAG
CCCTGCGAGAG
CCCTGCGGGAG
CCTTGCGGGAA
CCCTGCGGGAG
CCCTGCGCGAG
CTCTGCGCGAG
CCCTGCGCGAG
CTGATTGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATCGACTA
CTGATTGACTA
CTGATCGACTA
CTGATTGACTA
CTCTATACATC
CTCTACACATC
CTCTACACATC
CTCTACACATC
CTCTACACGTC
CTCTACACGTC
CTCTACACGTC
CTCTACACGTC
CTCTACACGTC
CTCTACACATC
CTCTACACGTC
CTGTACACGTC
CTCTACACATC
CTGTATACATC
SOCS2
Peye 's pa ch
abomasum
abomasum mucosa
ad enal gland
al eola mac ophage
biceps
b ain
b ain s em
caecum
ca diac en icle
ce ebellum
ce eb um
ce ix
colon
co pus lu eum
duodenum
epididymis
hea
hypo halamus
kidney
li e
longissimus do si
lung
mamma y gland
mesen e ic lymph node
omen um
o a ian ollicles
o a y
pi ui a y gland
placen a (including memb anes)
p escapula lymph node
ec um
enal co ex
enal medulla
umen
skin
skin om back
skin om side
spleen
es is
hy oid gland
onsil
u e us
en icle
whi e adipose
whole emb yo












































01.9
log10(FPKM)
Roslin - emale, adul
Roslin - emale, ju enile
Roslin - emale, adul
Jiang e al. (2014) - Texel
SOCS2
   
   
   
   
   
   
   
   
   




   
   
   
   
   
   
   
   
   
   
   
   

  
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
   
2945
410
275
49
(6)
4
(0.3)
67
(19)
49
(13)
Candida e Genes
Tibe an
Candida e Genes
High Al i ude
Candida e Genes
P ede ined gene panel
Z(FST)
A
B
C
D
E
F
2.0
1.0
SOCS2
FIG.3.Genomic egions wi h s ong selec i e signals in Tibe an sheep. (A) Dis ibu ion o log
2
(h
p
a ios) and Z(F
ST
) alues calcula ed in 100-kb
sliding windows wi h 50-kb inc emen s be ween Tibe an g oup (including b eeds ZNQ, ZCD, and ZRK om he pla eau en i onmen ) and con ol
g oup (including b eeds HUS and WDS om Eas China). The da a poin s in ed (co esponding o he op 5% o he empi ical log
2
(h
p
a ios) a io
dis ibu ion wi h alues >0.35 and he op 5% o he empi ical Z(F
ST
) dis ibu ion wi h alues >1.83) a e genomic egions unde selec ion in
Tibe an sheep. (B) Compa ison be ween he o e lap o candida e genes and he o e lap expec ed by chance. Numbe s in he in e sec ion egions
a e he obse ed o e lapping genes among he candida e genes in Tibe an sheep, sheep b eeds om he high-al i ude egions and he p ede ined
gene panel (i.e., he p e iously published candida e genes in o he mammalian species unde he high-al i ude en i onmen , including human,
dog, wol , yak, pig, and Tibe an an elope). Numbe s in pa en heses show he numbe o genes expec ed by chance. The o al numbe s o genes o
he sheep and he gene panel in ol ed in he es a e 18,013 and 65,029, espec i ely. (C) log
2
(h
p
a ios) and F
ST
alues a ound he SOCS2 locus. The
black and ed lines ep esen he log
2
(h
p
a ios) and F
ST
alues, espec i ely. (D) Tajima’s D alues a ound he SOCS2 locus. The blue and pu ple
lines ep esen he Tibe an sheep and con ol sheep, espec i ely. (E) E olu iona y analysis o he SOCS2 gene. The in e -species NJ ee is de i ed
om he 12 e eb a e o hologous sequences, and he mu a ions a e ma ked in ed. O. a ies M., namely O. a ise mu an , ep esen s Tibe an sheep
in which he mu a ions we e obse ed. O. a ies W., namely O. a ies wild, e e s o o he sheep b eeds in which he nucleo ides we e conse ed. (F)
Gene exp ession o SOCS2 in di e en sheep issues is based on ou di e en expe imen s deposi ed in he EBI Gene Exp ession A las da abase. The
FPKM ( agmen s pe kilobase o ansc ip pe million mapped eads) alue is used o measu e he exp ession le el.
Yang e al. .doi:10.1093/molbe /msw129 MBE
2582
en i onmen s; igs. 3Band 4B). Simila ly, he numbe o o e -
lapping candida e genes among he F
ST
,h
p
a io, XP-EHH, and
LFMM analyses signi ican ly exceeded hose expec ed by
chance ( igs. 5Cand 6C). These esul s p o ided u he sup-
po ha he candida e genes we e eliable.
Adap i e Mechanisms in Pla eau En i onmen s
Among he candida e genes o he pla eau adap a ions, i e
(IFNGR2,MAPK4,NOX4,SLC2A4,andPDK1) we e loca ed in
he classical HIF-1 (hypoxia-induced ac o s) pa hway, which
plays a cen al ole in egula ing cellula esponses o hypoxia
-4
-2
0
2
4
6
8
-4 -2 0 2 4
3
5
8
10
13
0 0.2 0.4
20
40
60
80
100
20 40 60 80 100
Cumula i e (%)
Cumula i e (%)
F equency (%)
F equency (%)
Z(FST)
log2(θπ·con ol/θπ·Taklimakan Dese )
GPX3
Peye 's pa ch
abomasum
abomasum mucosa
ad enal gland
al eola mac ophage
biceps
b ain
b ain s em
caecum
ca diac en icle
ce ebellum
ce eb um
ce ix
colon
co pus lu eum
duodenum
epididymis
hea
hypo halamus
kidney
li e
longissimus do si
lung
mamma y gland
mesen e ic lymph node
omen um
o a ian ollicles
o a y
pi ui a y gland
placen a (including memb anes)
p escapula lymph node
ec um
enal co ex
enal medulla
umen
skin
skin om back
skin om side
spleen
es is
hy oid gland
onsil
u e us
en icle
whi e adipose
whole emb yo
Roslin - emale, adul
Roslin - emale, ju enile
Roslin - emale, adul
Jiang e al. (2014) - Texel
4.4
log10(FPKM)
GPX3
Candida e Genes
Taklimakan Dese
Candida e Genes
A id and semi-a id
Candida e Genes
Bac ian camel
287 377
250
7
(3.6)
27
(5.3)
A
B
-1.0
0.0
1.0
2.0
3.0
-0.1
0.0
0.1
0.2
0.3
0.4
0.5
FST
log2(θπ·con ol/θπ·Taklimakan Dese )
Tajimas’D
-1.0
0.0
1.0
2.0
3.0
4.0
Taklimakan Dese sheep
Con ol sheep
FST
log2(θπ·con ol/θπ·Taklimakan Dese )
GPX3
C
D
E
F
0%50%100% 0% 50%100% 0% 50%100% 0% 50%100% 0% 50%100% 0% 50%100%
HUS
WDS
TCS
DQS
SPS
GZS
ZNQ
ZCD
ZRK
ZLZ
GDS
NMS
LOP
BRK
A
G
G
C
C
T
C
T
T
C
C
T
0
FIG.4.Genomic egions wi h s ong selec i e signals in sheep b eeds om he Taklimakan Dese egion. (A) Dis ibu ion o log
2
(h
p
a ios) and
Z(F
ST
) alues calcula ed in 100-kb sliding windows wi h 50-kb inc emen s be ween Taklimakan Dese g oup (including b eeds LOP and BRK om
he dese en i onmen ) and con ol g oup (including b eeds HUS and WDS om Eas China). The da a poin s in ed (co esponding o he op 5%
o empi ical log
2
(h
p
a ios) a io dis ibu ion wi h alues >0.28 and he op 5% o empi ical Z(F
ST
) dis ibu ion wi h alues >1.86) a e genomic
egions unde selec ion in he sheep b eeds om he Taklimakan Dese egion. (B) Compa ison be ween he o e lap o candida e genes and he
o e lap expec ed by chance. Numbe s in he in e sec ion egions a e he obse ed o e lapping genes among he candida e genes in sheep b eeds
om he Taklimakan Dese egion, sheep b eeds om he a id egions and he p ede ined gene panel (i.e., he p e iously published candida e
genes in o he mammalian species unde he a id en i onmen , including Bac ian camel). Numbe s in pa en heses show he numbe o genes
expec ed by chance. The o al numbe s o genes o sheep and Bac ian camel in ol ed in he es a e 18,013 and 20,251, espec i ely. (C) log
2
(h
p
a ios) and F
ST
alues a ound he GPX3 locus. The black and ed lines ep esen log
2
(h
p
a ios) and F
ST
alues, espec i ely. (D) Tajima’s D alues
a ound he GPX3 locus. The blue and pu ple lines ep esen he Taklimakan Dese sheep and con ol sheep, espec i ely. (E) Allele equencies o
six SNPs wi hin he GPX3 gene ac oss Chinese na i e sheep b eeds. The dese b eeds include BRK and LOP, whe eas he non-dese b eeds
comp ise all o he b eeds. (F) Gene exp ession o GPX3 in di e en sheep issues is based on ou di e en expe imen s deposi ed in he EBI Gene
Exp ession A las da abase. The FPKM ( agmen s pe kilobase o ansc ip pe million mapped eads) alue is used o measu e he exp ession le el.
Fo he abb e ia ions o he b eeds, see supplemen a y able S1,Supplemen a y Ma e ial online.
Rapid Adap a ion o Sheep o Ex eme En i onmen s .doi:10.1093/molbe /msw129 MBE
2583
(F ede and Fand ey 2013); ou we e ound in he co espon-
ding downs eam ascula endo helial g ow h ac o (VEGF;
RRAS and MAPK4) and glycolysis/gluconeogenesis (KIF2A
and KHSRP) pa hways; and i e (ARHGEF19,STK17A,
MAPK4,PPP1R1A and PRKG1)we e oundin he ascula
smoo h muscle con ac ion (VSMC) pa hway, which adjus s
he diame e o blood essels and he deli e y o blood oxy-
gen ( ig. 7A). We also ound nine posi i ely selec ed genes
(PLIN2,TWIST1,KIF2A,PTPN9,SOCS2,NCOA3,STK17A,
PDGFD,andLONP1) egula ing speci ic genes in he HIF-1
pa hway, wo genes (NF1 and HAND2) a ec ing he VEGF
signaling pa hway, one gene (LONP1) in luencing he glyco-
lysis/gluconeogenesis pa hway and wo genes (NOSIP and
SPSB4) egula ing he ni ic oxide (NO) compound in he
VSMC pa hway ( ig. 7A). The ne wo k o hese ele an pa h-
ways indica ed ha hypoxia-induced ac o s, angiogenesis,
asodila a ion and glycolysis me abolism a e he mos impo -
an ac o s ha allow sheep o manage ex eme hypoxic
en i onmen al p essu e. Fu he mo e, he unc ions o he
p oposed pa hway genes a e ela ed o he egula ion o
he ca dio ascula sys em and ene gy me abolism unde
hypoxic condi ions in human and mu ine cells (e.g., Mo a
e al. 2003;Tang e al. 2003;Kim e al. 2006;Jian e al. 2010;
Sawada e al. 2012;Pin i e al. 2015). Fo a de ailed desc ip ion
o he ele an unc ions o he pa hway candida e genes,
please see supplemen a y in o ma ion,Supplemen a y
Ma e ial online.
In addi ion, six candida e genes (FSTL1,PVR,EXT2,ALT4,
SOX6,andHAND2) anking wi hin he op 20 Z(F
ST
) alues
and wo addi ional candida e genes (PDGFD and BMPR2;
supplemen a y able S20,Supplemen a y Ma e ial online)
we e unc ionally in ol ed in bone, muscle, c anio acial,
limb, skin, and emb yonic de elopmen in e eb a es based
on he NCBI anno a ions and Ensembl da abases as well as
p e ious unc ional s udies in humans, mice and zeb a ish (e.
g., No on e al. 2005;Kayse ili e al. 2009;Geng e al. 2011;
Long e al. 2015;supplemen a y in o ma ion,Supplemen a y
Ma e ial online). These indings indica ed ha hese genes
may ha e played a dominan ole in egula ing body mo -
phology in Tibe an sheep, sugges ing ha selec ion on mo -
phology is an impo an adap i e mechanism unde he
pla eau en i onmen , which is consis en wi h he small
A
B
C
FIG.5.Analysis o he signa u es o posi i e selec ion in he genome o Tibe an sheep. Genomic landscape o he (A) XP-EHH alues and (B)
P- alues in he LFMM analysis in he genome o Tibe an sheep. The genes isualized in (A) and (B) a e he candida e genes om he signaling
pa hways o Tibe an sheep in ig. 7A.(C) Numbe o candida e genes iden i ied in Tibe an sheep by he ou me hods lis ed in each o he Venn
diag am componen s. Numbe s in pa en heses show he numbe o genes expec ed by chance.
Yang e al. .doi:10.1093/molbe /msw129 MBE
2584
Re e ences
Alkan C, Coe BP, Eichle EE. 2011. Genome s uc u al a ia ion disco e y
and geno yping. Na Re Gene . 12:363–376.
Ba e JC, F y B, Malle J, Daly MJ. 2005. Haplo iew: analysis and isua-
liza ion o LD and haplo ype maps. Bioin o ma ics 21:263–265.
B eye MD, B eye RM. 2000. P os aglandin ecep o s: hei ole in
egula ing enal unc ion. Cu Opin Neph ol Hype ens. 9:23–29.
B own ME, Funk CC. 2008. Food secu i y unde clima e change. Science
319:580–581.
Ca oll MA, Balazy M, Ma gio a P, Huang DD, Falck JR, McGi JC. 1996.
Cy och ome P-450-dependen HETEs: p o ile o biological ac i-
i y and s imula ion by asoac i e pep ides. Am J Physiol.
271:R863–R869.
Cho YS, Hu L, Hou H, Lee H, Xu J, Kwon S, Oh S, Kim H-M, Jho S, Kim S,
e al. 2013. The ige genome and compa a i e analysis wi h lion and
snow leopa d genomes. Na Commun. 4:2433.
de Magalh~
aes JP, Cos a J. 2009. A da abase o e eb a e longe i y eco ds
and hei ela ion o o he li e-his o y ai s. JE olBiol. 22:1770–1774.
Du L-X. 2011. Animal gene ic esou ces in China. Beijing: China
Ag icul u e P ess.
Eas e ling DR, Meehl GA, Pa mesan C, Changnon SA, Ka l TR, Mea ns
LO. 2000. Clima e ex emes: obse a ions, modeling, and impac s.
Science 289:2068–2074.
Ehle s J, Gibba d PL. 2007. The ex en and ch onology o Cenozoic global
glacia ion. Qua e n In . 164–165:6–20.
Eswo hy RS, Chu FF, Pax on RJ, Akman S, Do oshow JH. 1991.
Cha ac e iza ion and pa ial amino acid sequence o human plasma
glu a hione pe oxidase. A ch Biochem Biophys. 286:330–336.
F ede S, Fand ey J. 2013. Cellula and molecula de enses agains hypoxia.
In: Swenson ER, B€
a sch P, edi o s. High al i ude: human adap a ion
o hypoxia. New Yo k: Sp inge P ess. p. 26.
F eedman AH, G onau I, Schweize RM, O ega-Del Vecchyo D, Han E,
Sil a PM, Gala e ni M, Fan Z, Ma x P, Lo en e-Galdos B, e al. 2014.
Genome sequencing highligh s he dynamic ea ly his o y o dogs.
PLoS Gene 10:e1004016.
F eeman JL, Pe y GH, Feuk L, Redon R, McCa oll SA, Al shule DM,
Abu a aniH,JonesKW,Tyle -Smi hC,Hu lesME.2006.Copynumbe
a ia ion: new insigh s in genome di e si y. Genome Res. 16:949–961.
F icho E, Scho ille SD, Boucha d G, F anc¸ois O. 2013. Tes ing o asso-
cia ions be ween loci and en i onmen al g adien s using la en ac-
o mixed models. Mol Biol E ol. 30:1687–1699.
Ga alica Z, Lillebe g SL, Koul MS, Vanecek T, Hes O, Wang B, Michal M.
2008. COX-2 gene polymo phisms and p o ein exp ession in eno-
medulla y in e s i ial cell umo s. Hum Pa hol. 39:1495–1504.
GeR-L,CaiQ,ShenY-Y,SanA,MaL,ZhangY,YiX,ChenY,YangL,
Huang Y, e al. 2013. D a genome sequence o he Tibe an an e-
lope. Na Commun. 4:1858.
Geng Y, Dong Y, Yu M, Zhang L, Yan X, Sun J, Qiao L, Geng H, Nakajima
M, Fu uichi T, e al. 2011. Follis a in-like 1 (Fs l1) is a bone mo pho-
gene ic p o ein (BMP) 4 signaling an agonis in con olling mouse
lung de elopmen . P oc Na l Acad Sci USA. 108:7058–7063.
GouX,WangZ,LiN,QiuF,XuZ,YanD,YangS,JiaJ,KongX,WeiZ,e al.
2014. Whole-genome sequencing o six dog b eeds om con inuous
al i udes e eals adap a ion o high-al i ude hypoxia. Genome Res.
24:1308–1315.
Gu enkuns RN, He nandez RD, Williamson SH, Bus aman e CD. 2009.
In e ing he join demog aphic his o y o mul iple popula ions om
mul idimensional SNP equency da a. PLoS Gene . 5:e1000695.
Hancock AM, Cla k VJ, Qian Y, Di Rienzo A. 2011. Popula ion gene ic
analysis o he uncoupling p o eins suppo s a ole o UCP3 in
human cold esis ance. Mol Biol E ol. 28:610–614.
Ho mann I. 2010. Clima e change in con ex : implica ions o li es ock
p oduc ion and di e si y. In: Odongo NE, Ga cia M, Viljoen GJ,
edi o s. Sus ainable imp o emen o animal p oduc ion and heal h.
Vienna: IAEA-FAO. p. 33–44.
Howden SM, Soussana JF, Tubiello FN, Chhe i N, Dunlop M, Meinke H.
2007. Adap ing ag icul u e o clima e change. P oc Na l Acad Sci
USA. 104:19691–19696.
Jian B, Wang D, Chen D, Voss J, Chaud y I, Raju R. 2010. Hypoxia-induced
al e a ion o mi ochond ial genes in ca diomyocy es— ole o Bnip3
and Pdk1. Shock 34:169–175.
Jiang J-C, He M-L, Ka ma R, Penpa T. 1991. A compa a i e s udy on
se e al hema ologic alues o Tibe Pla eau sheep a di e en
al i udes. Chinese J Anim Sci. 27:11–14.
JiangY,XieM,ChenW,Talbo R,MaddoxJF,Fa au T,WuC,Muzny
DM, Li Y, Zhang W, e al. 2014. The sheep genome illumina es
biology o he umen and lipid me abolism. Science 344:1168–1173.
Ji imu uWZ,DingG,ChenG,SunY,SunZ,ZhangH,WangL,HasiS,
Zhang Y, Li J, e al. 2012. Genome sequences o wild and domes ic
Bac ian camels. Na Commun. 3:1202.
Kayse iliH,UzE,NiessenC,Va gelI,AlanayY,TuncbilekG,Yigi G,
Uygune O,CandanS,Oku H,e al.2009.ALX4 dys unc ion dis up s
c anio acial and epide mal de elopmen . Hum Mol Gene .
18:4357–4366.
Kijas JW, Lens a JA, Hayes B, Boi a d S, Ne o LRP, San C is obal M, Se in
B, McCulloch R, Whan V, Gie zen K, e al. 2012. Genome-wide
analysis o he wo ld’s sheep b eeds e eals high le els o his o ic
mix u e and s ong ecen selec ion. PLoS Biol. 10:e1001258.
Kim E-S, Elbel agy AR, Aboul-Naga AM, Rischkowsky B, Say e B,
Mwacha o JM, Ro hschild MR. 2016. Mul iple genomic signa u es
o selec ion in goa s and sheep indigenous o a ho a id en i on-
men . He edi y 116:255–264.
Kim JW, Tche nyshyo I, Semenza GL, Dang CV. 2006. HIF-1-media ed
exp ession o py u a e dehyd ogenase kinase: a me abolic swi ch
equi ed o cellula adap a ion o hypoxia. Cell Me ab. 3:177–185.
Ko neliussen TS, Alb ech sen A, Nielsen R. 2014. ANGSD: analysis o nex
gene a ion sequencing da a. BMC Bioin o ma ics 15:356.
Legge T. 1996. The beginnings o cap ine domes ica ion. In: Ha is DR,
edi o . The o igins and sp ead o ag icul u e and pas o alism in
Eu asia. New Yo k: Smi hsonian Ins i u ion P ess. p. 238–262.
Lehmkuhl F, Owen LA. 2005. La e Qua e na y glacia ion o Tibe and he
bo de ing moun ains: a e iew. Bo eas 34:87–100.
Li H, Du bin R. 2009. Fas and accu a e sho ead alignmen wi h
Bu ows-Wheele ans o m. Bioin o ma ics 25:1754–1760.
Li H, Du bin R. 2011. In e ence o human popula ion his o y om indi-
idual whole-genome sequences. Na u e 475:493–496.
Li H, Handsake B, Wysoke A, Fennell T, Ruan J, Home N, Ma h G,
Abecasis G, Du bin R. 2009. The sequence alignmen /map o ma
and SAM ools. Bioin o ma ics 25:2078–2079.
Li M, Tian S, Jin L, Zhou G, Li Y, Zhang Y, Wang T, Yeung CK, Chen L, Ma J,
e al. 2013. Genomic analyses iden i y dis inc pa e ns o selec ion in
domes ica ed pigs and Tibe an wild boa s. Na Gene . 45:1431–1438.
Lib ado P, De Sa kissian C, E mini L, Schube M, J
onsson H, Alb ech sen
A, Fumagalli M, Yang MA, Gamba C, Sequin-O lando A, e al. 2015.
T acking he o igins o Yaku ian ho ses and he gene ic basis o hei
as adap a ion o suba c ic en i onmen s. P oc Na l Acad Sci USA.
112:E6889–E6897.
Liu F-Y, Chen Q-H. 2006. E ec o acu e hypoxia on he pa ame e s o
blood gas in Tibe an Pla eau sheep. Chinese J Zool. 41:48–52.
LiuS,Lo enzenED,FumagalliM,LiB,Ha isK,XiongZ,ZhouL,
Ko neliussen TS, Somel M, Babbi C, e al. 2014. Popula ion geno-
mics e eal ecen specia ion and apid e olu iona y adap a ion in
pola bea s. Cell 157:785–794.
Lobell DB, Bu ke MB, Tebaldi C, Mas and ea MD, Falcon WP, Naylo RL.
2008. P io i izing clima e change adap a ion needs o ood secu i y
in 2030. Science 319:607–610.
LongL,O mis onML,YangX,Sou hwoodM,G
€
a S, Machado RD,
Muelle M, Kinzel B, Yung LM, Wilkinson JM, e al. 2015. Selec i e
enhancemen o endo helial BMPR-II wi h BMP9 e e ses pulmona y
a e ial hype ension. Na Med. 21:777–785.
Lo enzen ED, Nogue´s-B a o D, O lando L, Weins ock J, Binladen J,
Ma ske KA, Ugan A, Bo egaa d MK, Gilbe MT, Nielsen R, e al.
2011. Species-speci ic esponses o La e Qua e na y mega auna o
clima e and humans. Na u e 479:359–364.
L F-H, Agha S, Kan anen J, Colli L, S ucki S, Kijas JW, Joos S, Li M-H,
Ajmone Ma san P. 2014. Adap a ions o clima e-media ed selec i e
p essu es in sheep. Mol Biol E ol. 31:3324–3343.
Rapid Adap a ion o Sheep o Ex eme En i onmen s .doi:10.1093/molbe /msw129 MBE
2591

L F-H, Peng W-F, Yang J, Zhao Y-X, Li W-R, Liu M-J, Ma Y-H, Zhao Q-J,
Yang G-L, Wang F, e al. 2015. Mi ogenomic me a-analysis iden i ies
wo phases o mig a ion in he his o y o eas e n Eu asian sheep. Mol
Biol E ol. 32:2515–2533.
MaT,WangJ,ZhouG,YueZ,HuQ,ChenY,LiuB,QiuQ,WangZ,
Zhang J, e al. 2013. Genomic insigh s in o sal adap a ion in a dese
popla . Na Commun. 4:2797.
Mackelp ang R, Wald op MP, DeAngelis KM, Da id MM, Cha a ia KL,
Blazewicz SJ, Rubin EM, Jansson JK. 2011. Me agenomic analysis o a
pe ma os mic obial communi y e eals a apid esponse o haw.
Na u e 480:368–371.
Maqui e JJ, Johnson CM, Mock idge JW, Da enpo AP. 1997. Endo helin
con e ing enzyme (ECE) ac i i y in human ascula smoo h muscle.
B J Pha macol. 122:1647–1654.
Mi H, Mu uganujan A, Casag ande JT, Thomas PD. 2013. La ge-scale
gene unc ion analysis wi h he PANTHER classi ica ion sys em.
Na P o oc. 8:1551–1566.
Mi H, Poudel S, Mu uganujan A, Casag ande JT, Thomas PD. 2016.
PANTHER e sion 10: expanded p o ein amilies and unc ions,
and analysis ools. Nucleic Acids Res. 44:D336–D342.
Mo aA,Da iesAM,Be andL,Sha i I,BudasGR,Jo ano i

cS,Mou on
V, Kahn CR, Lucocg JM, G ay GA, e al. 2003. De iciency o PDK1 in
ca diac muscle esul s in hea ailu e and inc eased sensi i i y o
hypoxia. Embo J. 22:4666–4676.
Mys e ud A, S ense h NC, Yoccoz NG, Lang a n R, S einheim G. 2001.
Nonlinea e ec s o la ge-scale clima ic a iabili y on wild and do-
mes ic he bi o es. Na u e 410:1096–1099.
Mys e ud A, Yoccoz NG, Lang a n R, Pe o elli N, S ense h NC. 2008.
Hie a chical pa h analysis o dee esponses o di ec and indi ec
e ec s o clima e in no he n o es . Philos T ans R Soc Lond B Biol
Sci. 363:2359–2368.
Nielsen R, Ko neliussen T, Alb ech sen A, Li Y, Wang J. 2012. SNP calling,
geno ype calling, and sample allele equency es ima ion om new-
gene a ion sequencing da a. PLoS One 7:e37558.
No on WHJ, Ledin J, G andel H, Neumann CJ. 2005. HSPG syn hesis by
zeb a ish Ex 2 and Ex l3 is equi ed o Fg 10 signalling du ing limb
de elopmen . De elopmen 132:4963–4973.
Oakesho JG, Wilson SR, Knibb WR. 1988. Selec ion a ec ing enzyme
polymo phisms in enclosed D osophila popula ions main ained in a
na u al en i onmen . P oc Na l Acad Sci USA. 85:293–297.
Oleksyk TK, Smi h MW, O’B ien SJ. 2010. Genome-wide scans o oo p in s
o na u al selec ion. Philos T ans R Soc Lond B Biol Sci. 365:185–205.
O lando L, Ginolhac A, Zhang G, F oese D, Alb ech sen A, S ille M,
Schube M, Cappellini E, Pe e sen B, Mol ke I, e al. 2013.
Recalib a ing Equus e olu ion using he genome sequence o an
ea ly Middle Pleis ocene ho se. Na u e 499:74–78.
Pa k SD, Magee DA, McGe igan PA, Teasdale MD, Edwa ds CJ, Lohan
AJ, Mu phy A, B aud M, Donoghue MT, Liu Y, e al. 2015. Genome
sequencing o he ex inc Eu asian wild au ochs, Bos p imigenius,
illumina es he phylogeog aphy and e olu ion o ca le. Genome Biol
16:234.
Pa ke KL, Robbins CT. 1985. The mo egula ion in ungula es. In: Hudson
RJ, Whi e RG, edi o s. Bioene ge ics o wild he bi o es. Boca Ra on,
FL: CRC P ess, Inc. p. 161–182.
Pa e son N, P ice AL, Reich D. 2006. Popula ion s uc u e and eigena-
nalysis. PLoS Gene . 2:e190.
Pe yszak R, Bu de T, Fio elli B, Fonseca NA, Gonzalez-Po a M,
Has ings E, Hube W, Jupp S, Keays M, K y ych N, e al. 2014.
Exp ession A las upda e—a da abase o gene and ansc ip exp es-
sion om mic oa ay- and sequencing-based unc ional genomics
expe imen s. Nucleic Acids Res. 42:D926–D932.
Piao S, Ciais P, Huang Y, Shen Z, Peng S, Li J, Zhou L, Liu H, Ma Y, Ding Y,
e al. 2010. The impac s o clima e change on wa e esou ces and
ag icul u e in China. Na u e 467:43–51.
Pick ell JK, P i cha d JK. 2012. In e ence o popula ion spli s and mix u es
om genome-wide allele equency da a. PLoS Gene . 8:e1002967.
Pin i M, Gibellini L, Liu Y, Xu S, Lu B, Cossa izza A. 2015. Mi ochond ial
Lon p o ease a he c oss oads o oxida i e s ess, aging and cance .
Cell Mol Li e Sci. 72:4807–4824.
Pu cell S, Neale B, Todd-B own K, Thomas L, Fe ei a MA, Bende D,
Malle J, Skla P, de Bakke PI, Daly MJ, e al. 2007. PLINK: a ool se o
whole-genome associa ion and popula ion-based linkage analyses.
Am J Hum Gene . 81:559–575.
Qiu Q, Zhang G, Ma T, Qian W, Wang J, Ye Z, Cao C, Hu Q, Kim J, La kin
DM, e al. 2012. The yak genome and adap a ion o li e a high
al i ude. Na Gene . 44:946–949.
Qu Y, Zhao H, Han N, Zhou G, Song G, Gao B, Tian S, Zhang J, Zhang R,
Meng X, e al. 2013. G ound i genome e eals a ian adap a ion o
li ing a high al i udes in he Tibe an pla eau. Na Commun. 4:2071.
Rico-Bau is a E, Flo es-Mo ales A, Fe n
andez-Pe´ ez L. 2006. Supp esso o
cy okine signaling (SOCS) 2, a p o ein wi h mul iple unc ions.
Cy okine G ow h Fac o Re . 17:431–439.
Sabe i PC, Va illy P, F y B, Lohmuelle J, Hos e e E, Co sapas C, Xie X,
By ne EH, McCa oll SA, Gaude R, e al. 2007. Genome-wide de ec-
ion and cha ac e iza ion o posi i e selec ion in human popula-
ions. Na u e 449:913–918.
SawadaJ,U akamiT,LiF,U akamiA,ZhuW,FukudaM,LiDY,Ruoslah i
E, Koma su M. 2012. Small GTPase R-Ras egula es in eg i y and
unc ionali y o umo blood essels. Cance Cell 22:235–249.
Schein eld LB, Soi S, Thompson S, Rancia o A, Woldemeskel D,
Beggs W, Lambe C, Ja is JP, Aba e D, Belay G, e al. 2012.
Gene ic adap a ion o high al i ude in he E hiopian highlands.
Genome Biol. 13:R1.
Sezgin E, Du e nell DD, Ma zkin LM, Duan Y, Zhu CT, Ve elli BC, Eanes
WF. 2004. Single locus la i udinal clines and hei ela ionship o
empe a e adap a ion in me abolic genes and de i ed alleles in
D osophila melanogas e . Gene ics 168:923–931.
SimonsonTS,YangY,Hu CD,YunH,QinG,Wi he spoonDJ,BaiZ,
Lo enzo FR, Xing J, Jo de LB, e al. 2010. Gene ic e idence o high-
al i ude adap a ion in Tibe . Science 329:72–75.
S opka T, Zi ny JH, S opko a P, P chal JF, P chal JT. 1998. Human
hema opoie ic p ogeni o s exp ess e y h opoie in. Blood 91:3766–3772.
Tajima F. 1989. S a is ical me hod o es ing he neu al mu a ion hy-
po hesis by DNA polymo phism. Gene ics 123:585–595.
Tang H, Peng J, Wang P, Risch NJ. 2005. Es ima ion o indi idual admix-
u e: analy ical and s udy design conside a ions. Gene Epidemiol.
28:289–301.
TangM,WangGR,LuP,Ka asRH,A ono i zM,Hexime SP,
Kal enb onn KM, Blume KJ, Side o ski DP, Zhu Y, e al. 2003.
Regula o o G-p o ein signaling-2 media es ascula smoo h muscle
elaxa ion and blood p essu e. Na Med. 9:1506–1512.
Tu nley AM. 2005. Role o SOCS2 in g ow h ho mone ac ions. T ends
Endoc inol Me ab. 16:53–58.
WangM-S,LiY,PengM-S,ZhongL,WangZ-J,LiQ-Y,TuX-L,DongY,
Zhu C-L, Wang L, e al. 2015. Genomic analyses e eal po en ial
independen adap a ion o high al i ude in Tibe an chickens. Mol
Biol E ol. 32:1880–1889.
Wang X, Ma YH, Chen H, Guan WJ. 2007. Gene ic and phylogene ic
s udies o Chinese na i e sheep b eeds (O is a ies) based on m DNA
D-loop sequences. Small Ruminan Res. 72:232–236.
WeiC,WangH,LiuG,WuM,CaoJ,LiuZ,LiuR,ZhaoF,ZhangL,LuJ,
e al. 2015. Genome-wide analysis e eals popula ion s uc u e and
selec ion in Chinese indigenous sheep b eeds. BMC Genomics 16:194.
Wei BS, Cocke ham CC. 1984. Es ima ing F-s a is ics o he analysis o
popula ion s uc u e. E olu ion 38:1358–1370.
WuH,GuangX,Al-FageehMB,CaoJ,PanS,ZhouH,ZhangL,
Abu a boush MH, Xing Y, Xie Z, e al. 2014. Camelid genomes e eal
e olu ion and adap a ion o dese en i onmen s. Na Commun.
5:5188.
ZhaoZ,XuD,WangL,HaoJ,WangJ,ZhouX,WangW,QiuQ,HuangX,
Zhou J, e al. 2016. Con e gen e olu ion o umen mic obiomes in
high-al i ude mammals. Cu Biol. 26(14):1873–1879.
Zhao E, Yu Q, Zhang N, Kong D, Zhao Y. 2013. Mi ochond ial DNA
di e si y and he o igin o Chinese indigenous sheep. T op Anim
Heal h P o. 45:1715–1722.
Zheng B, Xu Q, Shen Y. 2002. The ela ionship be ween clima e change
and Qua e na y glacial cycles on he Qinghai-Tibe an Pla eau: e iew
and specula ion. Qua e n In . 97–98:93–101.
Yang e al. .doi:10.1093/molbe /msw129 MBE
2592