Riverzzz13com

NatureGeneticsVOLUME45|NUMBER5|MAY2013501ArticlesApproachesexploitingtraitdistributionextremesmaybeusedtoidentifylociassociatedwithcommontraits,butitisunknownwhethertheselociaregeneralizabletothebroaderpopulation.
Inagenome-widesearchforlociassociatedwiththeupperversusthelower5thpercentilesofbodymassindex,heightandwaist-to-hipratio,aswellasclinicalclassesofobesity,includingupto263,407individualsofEuropeanancestry,weidentified4newloci(IGFBP4,H6PD,RSRC1andPPP2R2A)influencingheightdetectedinthedistributiontailsand7newloci(HNF4G,RPTOR,GNAT2,MRPS33P4,ADCY9,HS6ST3andZZZ3)forclinicalclassesofobesity.
Further,wefindalargeoverlapingeneticstructureandthedistributionofvariantsbetweentraitsbasedonextremesandthegeneralpopulationandlittleetiologicalheterogeneitybetweenobesitysubgroups.
Althoughitispossiblethatothergeneticorenvironmentalfactorsmod-ifythemanifestationsofthesevariants,producinganextremephenotypeonlyinselectedindividuals,itisalsoconceivablethatthetraitextremesare,atleastinpart,etiologicallydistinctanddifferentfromthoseactinginthegeneralpopulation.
Withintheextremesofthetraitdistribution,theremaybeetiologicallydiscretesubgroupsorenrichmentforlesscommoncausalvariants19.
Althoughanalyzingthefulldistributionisgenerallymorepowerful,incaseswherethereisheterogeneity,analyzingextremesbycase-controldesignmayoffersuperiorpower29.
Theextremesforanthropometrictraits,particularlyBMI,havebeendefinedinnumerousways,includingusingthetailsofthefullpopulationdistribution(forexample,>95thor>97thpercentile)andabsolutecutoffs(forexample,≥40kg/m2)basedonclinicalorstandardreferences,andsomestudieshaveusedacombinationofdefinitionsfortheirdiscoveryandreplicationanalyses.
Thecommondenominatorforstudiesaddressingtraitextremes(hereinusedasamoregenericterm)isthattheydichotomizethetraitdistributionandanalyzeddatausingacase-controldesign.
Studiessuggestthatthepercentilecutoffchoiceandascertainmentstrategyusedmayaffecttheobservedriskandsubsequentpower30,31;however,theconse-quencesofthedefinitionsoftraitextremesonthediscoveryandchar-acterizationoflociforcomplextraitshavenotbeensystematicallyevaluated.
Inthepresentstudy,wehaveusedtheterms'distributiontails'todescribeanalysescomparingtheupperandlower5thpercen-tilesofthetraitdistributions;'clinicalclassesofobesity'todescribeanalyseswherecontrolsweresubjectswithBMI0.
007forallSNPs,Bonferronicorrected).
Fouroutofsevennewobesity-relatedlocishowedsignificanceatP<0.
007(Bonferronicorrected)inthesestudiesofextremelyobeseindividuals.
EffectsofknownextremeobesitylociinourstudyPreviousstudiesofextremechildhoodand/oradultobesityusingdifferentascertainmentstrategieshavereportedgenome-widesig-nificantorneargenome-widesignificantassociations(P<5*107)Table1Newlocireachinggenome-widesignificance(P<5*108)forthetailsofanthropometrictraitsandclinicalclassesofobesityStage1Stage2aStage1+stage2SNPChr.
PositionNearbygeneEffectalleleOtheralleleEffectallelefreq.
Cases(n)Controls(n)ORPCases(n)Controls(n)ORPCases(n)Controls(n)ORPHeighttailsrs5844381735852698IGFBP4CA0.
6177,8307,8501.
181.
11*1091,8141,8141.
190.
0019,6449,6641.
185.
22*1012rs666250919240191H6PDTC0.
14635,4625,4611.
232.
21*1063,6153,5661.
233.
37*1059,0779,0271.
233.
19*1010rs23629653159592073RSRC1-SHOX2TA0.
507,9897,9931.
141.
45*1074,8194,7751.
100.
00212,80812,7681.
122.
14*109rs1594829826261994PPP2R2ACT0.
76886,6936,6971.
185.
51*1074,1664,1151.
110.
0110,85910,8121.
153.
88*108ObesityclassIIrs79893361395815549HS6ST3AG0.
47049,82562,1141.
125.
88*1091,66417,1131.
040.
2511,48979,2261.
101.
06*108rs17381664177820919ZZZ3CT0.
39239,83362,1141.
117.
61*1085,35133,8411.
050.
0415,18495,9551.
092.
85*108ObesityclassIrs170242581109948844GNAT2TC0.
036418,66238,4271.
231.
41*1068,95615,4711.
281.
12*10627,61853,8981.
258.
66*1012rs4735692876778218HNF4GAG0.
583432,67565,6971.
075.
03*10822,08638,3521.
040.
00554,761104,0491.
062.
48*109rs130411262050526403MRPS33P4TC0.
717932,02064,0151.
073.
05*10722,08837,5951.
040.
00754,108101,6101.
062.
16*108rs2531995163953468ADCY9TC0.
614632,43365,5421.
063.
17*1066,68016,6021.
070.
00439,11382,1441.
074.
04*108Overweightclassrs4735692876778218HNF4GAG0.
583992,70365,6981.
056.
13*10965,32339,2901.
030.
003158,026104,9881.
043.
51*1010rs75038071776205706RPTORAC0.
565492,85565,7231.
044.
20*10664,53538,8131.
030.
0009157,390104,5361.
041.
98*108Chr.
,chromosome;freq.
,frequency;OR,oddsratio.
aStage2consistsofstudieswitheitherGWASorMetabochipdata.
NotallSNPswerepresentontheMetabochip.
504VOLUME45|NUMBER5|MAY2013NatureGeneticsArticleswithFTO,MC4R,TMEM18,FAIM2,TNKS,HOXB5,OLFM4,NPC1,MAF,PTER,SDCCAG8,PCSK1(rs6235andrs6232)andKCNMA1(refs.
14–16,22–26).
WiththeexceptionofPCSK1(rs6232)forthedistributiontailsofBMIandMAFforthedistributiontailsofBMIandobesityclassII,allassociationsshowedconsistentdirectionsofeffectacrosstheBMI-relatedoutcomes(SupplementaryTable11).
Ofthe13loci,replicationatasignificancelevelofP<0.
004(Bonferronicorrected)wasobservedfor4SNPs(FTO,MC4R,TMEM18andFAIM2)forthedistributiontailsofBMIandallclinicalclassesofobesity.
Twoloci,MAFandKCNMA1,whichhavethusfaronlybeenreportedforextremeobesity,werenotsignificantlyassociatedwithanyofourtraitsateitheraBonferroni-correctedornominalsignificancethreshold(P<0.
05).
EmpiricalpowercomparisonofthepopulationextremesandthefulldistributionIfthetraitextremeshavedifferentgeneticinheritanceorareetiologi-callymorehomogenousthanthefulldistribution,analyzingextremesortailsofthedistributionbycase-controldesignmayoffersuperiorpower.
Totestthisempirically,weconductedmeta-analysesofthefulldistributionsofBMIandheightwithallstudiesincludedinstage1andstage2.
Onlytwoloci(IGFBP4andH6PD)outofthefournewlociforthedistributiontailsofheightreachedgenome-widesignifi-cance(P<5*108)usingthefullheightdistribution(Table2).
Fourloci(GNAT2,ZZZ3,HNF4GandRPTOR)outofthesevennewlociidentifiedfortheclinicalclassesofobesityachievedgenome-widesignificanceforthefullBMIdistribution.
TheremaininglocihadPvaluesof<5*105inthefulldistributionand,thus,wouldlikelyhavebeendetectedwithalargersamplesize.
GeneticarchitectureinthedistributiontailsandfulldistributionToinvestigatedifferencesingeneticarchitecturebetweenthedistribu-tiontailsandthefulldistributions,weestimatedwhethertheobservedgeneticeffectsinthedistributiontailsofBMI,heightandWHRweredifferentfromwhatwouldbeexpectedbasedonthefulldistributionsofthecorrespondingtraits.
Todothis,wefirstestimatedtheexpectedeffectforeachSNPinthedistributiontailsonthebasisofthefulldistributionineachstudyandthencarriedoutmeta-analysisoftheexpectedassociationsacrossstudies.
Thequantile-quantileplotsofPvaluestestingdifferencesbetweentheobservedandexpectedeffects(Fig.
1andSupplementaryFig.
9)didnotshowanyenrichment,6655443322Expected–log10(P)Observed–log10(P)1100Figure1Quantile-quantileplotofthelog10PvaluesforthedifferencebetweentheobservedassociationforthedistributiontailsofBMIandtheexpectedassociationbasedontheoverallBMIdistribution.
Table2AssociationresultsfornewSNPsassociatedwithheight-andobesity-relatedtraitsatgenome-widesignificance(P<5*108)BMItailsObesityclassIIIObesityclassIIObesityclassIOverweightclassBMI(continuous)aHeighttailsHeight(continuous)aSNPGeneEffectalleleOtheralleleORPORPORPORPORPEffectPORPEffectPHeighttailsrs584438IGFBP4CA0.
980.
521.
020.
641.
010.
471.
000.
751.
000.
590.
0050.
221.
185.
22*10120.
0259.
43*1011rs6662509H6PDTC1.
000.
951.
110.
071.
010.
830.
990.
340.
990.
350.
0060.
271.
233.
19*10100.
0317.
76*1012rs2362965RSRC1-SHOX2TA0.
950.
020.
970.
250.
980.
200.
990.
370.
990.
210.
0070.
051.
122.
14*1090.
0177.
07*108rs1594829PPP2R2ACT1.
030.
331.
060.
111.
010.
581.
000.
821.
010.
320.
0040.
331.
153.
88*1080.
0164.
29*105ObesityclassIIrs7989336HS6ST3AG1.
090.
00011.
110.
00061.
101.
06*1081.
049.
38*1051.
042.
33*1060.
0168.
80*1061.
000.
890.
0010.
71rs17381664ZZZ3CT1.
080.
00011.
125.
41*1051.
092.
85*1081.
056.
80*1081.
042.
23*1070.
0222.
50*10111.
060.
0050.
0100.
004ObesityclassIrs17024258GNAT2TC1.
270.
021.
450.
0021.
267.
73*1051.
258.
66*10121.
131.
41*1080.
0674.
34*10141.
210.
080.
0100.
36rs4735692HNF4GAG1.
091.
97*1051.
080.
0061.
050.
00051.
062.
48*1091.
043.
51*10100.
0199.
94*10101.
020.
500.
0060.
10rs13041126MRPS33P4TC1.
080.
0011.
050.
161.
060.
00021.
062.
16*1081.
041.
43*1060.
0178.
52*1071.
020.
560.
0020.
65rs2531995ADCY9TC1.
060.
011.
090.
0061.
060.
0011.
074.
04*1081.
035.
57*1050.
0216.
58*1081.
070.
0050.
0181.
87*106Overweightclassrs4735692HNF4GAG1.
091.
97*1051.
080.
0061.
050.
00051.
062.
48*1091.
043.
51*10100.
0199.
94*10101.
020.
500.
0060.
10rs7503807RPTORAC1.
087.
07*1051.
139.
44*1061.
072.
46*1061.
051.
12*1071.
041.
98*1080.
0203.
00*10100.
990.
550.
0010.
85aTheβvaluerepresentsthedifferenceinstandardizedeffects.
NatureGeneticsVOLUME45|NUMBER5|MAY2013505Articlesindicatingthattheeffectsizesobservedindistributiontailsandthoseexpectedbasedontheoveralldistributionweresimilar.
Further,com-parableresultswereobservedforthe32SNPspreviouslyassociatedwithBMIinSpeliotesetal.
4,aswellasforpreviouslypublishedandnewlociforextremeobesity(SupplementaryTable12).
Tofurthercomparegeneticinheritanceinthedistributiontailswiththatinthefulldistribution,weusedapolygenicapproach45.
Themeta-analysisresultsofthedistributiontailsandfulldistributionwereusedtocreatetwopolygeneticscores(bysummingthenumberofrisk-associatedallelesateachSNP)insixstudies(SupplementaryTable13).
WefoundthatthepolygenicscorebasedonthefullBMIdistributionconsistentlyexplainedmoreofthevariancethanthescorebasedonthedistributiontails(forexample,15.
3%versus6.
4%atP<0.
05)(Fig.
2andSupplementaryTable14).
SimilarresultswereobservedforheightandWHR(SupplementaryFig.
10).
Onliabilityscale,thevarianceexplainedbythetwopolygenicscoreswassimilarfordifferentBMI-relatedoutcomes(SupplementaryFig.
11)anddifferentpercentilecutoffsusedtodefinethedistributiontails(datanotshown),suggestingthatthefractionoftheoverallvarianceexplainedbySNPsisnotinfluencedbyoutcomecategorizationbutbytheabilitytoaccuratelyrankandestimatetheβcoefficientsoftheassociation,whichisbetterachievedbyusingtheentirestudypopulationinsteadofthedistributiontails.
Ourresultsalsoindicatethatgeneticdeterminantsforthedistributiontailsaresimilartothoseforthefulldistributionandthatcommonvariantlocicontributetoextremephenotypes.
However,itshouldbenotedthatouranalysesoftheupperandlower5thpercentilesofthedistribution(tails)doesnotnecessarilyextendtomoreextremecutoffs,suchasthetopandbottom1stpercentiles.
AllelicheterogeneityatnewandpreviouslyidentifiedlociToexploreenrichmentforallelicheterogeneityinthedistribu-tiontailsandclinicalclassesofobesity,weperformedconditionalanalysesusingarecentlydescribedmethod46.
Intheseanalyses,wefoundsecondarysignalsthatreachedgenome-widesignificance(P<5*108)at17loci,including1locusforthedistributiontailsofBMI(FTO),13lociforthedistributiontailsofheight(PTCH1(2signals),GHSR,EDEM2,C6orf106,CRADD,EFEMP1,HHIP,FBXW11,NPR3,LINC00471(alsoknownasC2orf52),BCKDHBandEFR3B),1locusforthedistributiontailsofWHR(RSPO3),2locifortheoverweightclass(MC4RandFANCL)and1locusforobesityclassI(FANCL)(SupplementaryTable15).
WhereasthesecondarysignalsforthedistributiontailsofBMI(FTO)andWHR(RSPO3)andtheoverweightclassandobesityclassI(FANCL)havenotbeenestablishedpreviously,all13height-relatedlociidentifiedhere,aswellastheMC4Rlocus,havepreviouslybeenshowntohaveallelicheterogeneityinthegeneralpopulation7,9,suggestingthatthereisnoenrichmentinthedistributiontailsforsecondarysignals(SupplementaryFigs.
12–14).
Wealsolookedforevidenceofenrichmentofunobservedlow-frequencyvariantsbyconductinghaplotypeanalyseswithinknownandnewloci,ashaplotypesconstructedfromcommonSNPsmaytaglow-frequencyvariantsthatareenrichedinthetailsofthetraitdistributionsbutarerarerinthegeneralpopulation.
Usinggeno-typedatafromthelargeststudies,threesignalsofassociationwereobservedforthedistributiontailsofheightthatexceededconserva-tiveprioroddsofassociationof1in30,000:ID4(Bayesfactorof118,839),LIN28B(Bayesfactorof105,478)andDLEU7(Bayesfactorof66,599)(SupplementaryTable16).
However,forallthreeloci,associationsignalswerecharacterizedbytwoclustersofhaplotypes(bothcommonandrare)andwerenotconsistentwithenrichmentofunobservedlow-frequencycausalvariantsinthedistributiontails.
DISCUSSIONInourmeta-analysisofGWASofupto263,407individualsofEuropeanancestry,weidentified165lociassociatedwithdistribu-tiontails(theupperversuslower5thpercentiles)ofBMI,heightandWHRand/orclinicalclassesofobesity.
Elevenoftheselocihavenotpreviouslybeenassociatedwithanthropometrictraits.
Severalofthenewlociwerelocatednearstrongbiologicalcandidategenes,suchasIGFBP4andSHOX2forthedistributiontailsofheightandHNF4GandADCY9fortheoverweightclassand/orobesityclassI,suggest-ingfutureareasofresearch.
Althoughbyusingdifferentdistributioncutoffswediscoveredadditionallocithatwouldnothavebeenidenti-fiedasgenome-widesignificantusingthefulldistributionofthesamestudysamples,thereisnoevidencetosuggestthattheclinicalclassesofobesityareetiologicallydistinct,andthemajorityofevidenceindi-catesthatthepopulationextremessharemanyofthesamelociwiththegeneralpopulation.
Toassesstheimpactofdifferentdistributioncutoffsongeneticvariantsassociatedwiththepopulationextremes,wechosetoevalu-atethe5%tailsoftraitdistributionandclinicalclassesofobesity,specificallyobesityclassesIIandIII.
Althoughothershaveascer-tainedpopulationextremesdifferently,allvariantsassociatedwithobesity-relatedtraitsinourmeta-analysiswerefoundtohavedirec-tionallyconsistentresultsinfiveindependentstudiesofextremelyFigure2Varianceinextremeobesityexplainedbycommongeneticvariants.
ThephenotypicvarianceexplainedishigherwhenSNPswithlowerdegreesofsignificanceareincludedinthepolygeneticpredictionmodel.
Theyaxisrepresentstheproportionofvarianceexplained(NagelkerkeR2)forextremeobesityinsixstudiesnotincludedinthediscoverymeta-analysis.
Thethickerlinesrepresenttheweightedaverage;95%confidenceintervalsarereportedasdouble-headedarrows.
(a)Thepredictionmodelwasbasedontheresultsfromthestage1meta-analysisofthedistributiontailsofBMI.
(b)ThepredictionmodelwasbasedonBMIfromthefulldistribution(modifiedversionofthepreviousGIANTmeta-analysisbySpeliotesetal.
4).
TheEssenObesityStudywasnotadjustedbyage.
0.
25TailsofBMIabBMIMeta–analysisTwingeneLifeLinesEssenObesityStudyGEO–ITGOYAFrenchObesityStudy0.
200.
150.
10Varianceexplained(R2)0.
0500.
250.
200.
150.
100.
0505*10–85*10–75*10–65*10–55*10–45*10–30.
05Pvalues0.
10.
20.
40.
615*10–85*10–75*10–65*10–55*10–45*10–30.
05Pvalues0.
10.
20.
40.
61506VOLUME45|NUMBER5|MAY2013NatureGeneticsArticlesobesesamples.
Ofthe13locipreviouslyidentifiedasassociatedwithextremeobesity14–16,22–26,nearlyall(exceptPCSK1(rs6232)andMAF)showedaconsistentdirectionofeffectinthedistributiontailsofBMI.
Onlytwoloci(MAFandKCNMA1),originallyidentifiedforearly-onsetandmorbidadultobesity14,26,didnotreplicateforanyofourBMI-relatedoutcomes.
Althoughitispossiblethatwehadinsufficientpoweriftherewasasubstantialwinner'scursepresentintheinitialpublications,itisalsoconceivablethatthesesusceptibilitylociarepopulationspecific,onlycontributetoriskatyoungerages47,representfalsepositivefindingsortagrarecausalvariantsthataredifficulttodetectinpopulation-basedsamples.
BecauseourstudywasbasedonGWASdata,wewerenotwellsuitedtoaddresstheroleofrarevariantsinextremetraits.
Althoughhaplotype-basedanalysesidentifiedstrongassociationsofhaplotypesinthreegeneswiththedistributiontailsofheight,whichcouldsug-gestthattheyaretaggedbyrarevariants,suchputativevariantscouldnotbeestablishedusingourapproach.
Thesuggestionthatrarevari-antscouldbemoreimportantinthedistributionextremesofcomplextraitsneedstobeaddressedusingotherstudydesigns,suchasrese-quencingprojectsorusingthenewExomeChipmicroarraysthatarecurrentlybeinganalyzedinmanylargestudysamples.
Oursystematiccomparisonsbetweendistributionextremesandthefulldistributionyieldedseveralimportantinsightsthatalsomaybeinformativeforothercomplextraits.
Whencomparingobservedgeneticeffectsindistributiontailswiththeexpectedeffectsextrapo-latedfromtheoveralldistributionsofthecorrespondingtraits,wedidnotobserveanysystematicdifferences.
Further,weshowedthatthepolygenicscorebasedonthefulldistributionexplainedalargerproportionofvariancethanthescorebasedonthedistributiontails.
Takentogetherwiththefindingthathalfofournewlociwereassoci-atedatagenome-widesignificantlevelintheoveralldistribution,thisimpliesthatthereislimitedetiologicalheterogeneityintheseanthropometrictraits.
Ouranalysisshowsthat,whereassomecom-monvariantscanhavelargereffectsinthedistributionextremes,theseeffectsasawholearenotlargerthanexpectedbasedontheireffectsintheoveralldistribution.
Further,whereasrarevariantsspe-cifictothedistributionextremesmaystillexist,theextremessharemostofthecommonlociwiththeoveralldistribution.
Conclusionsthatcanbedrawnfromtheseobservationsarethat,whenaccessisavailabletodataforthefulldistribution,case-controlanalysesusingpopulationextremescanbeusefultofindadditionalloci.
Althoughanalyzingthefulldistributionisgenerallymorepowerful,smallamountsofheterogeneityinthedistributionmayallowfortheidentificationofadditionallocibyanalyzingthedatausingdifferentcutoffs,suchasthedistributiontails.
Further,asinmostcaseswhenresourcesarelimited,ourresultsindicatethatastrategywiththeselec-tionofindividualsfromthepopulationextremesforgeneticanalysescouldbeacost-effectiveapproachandwilllikelyyieldlocithatarerel-evantandcanlargelybeextendedtothegeneralpopulation.
Compatiblewiththefindingsfromrecent,smallerstudies21–23,ourresultsshowthatthistheoreticallyappealingapproachalsoholdsempirically.
Inconclusion,inourlargeGWASmeta-analysisincludingupto263,407individuals,weidentified4newlociinfluencingheightdetectedatthedistributiontails,aswellas7newlociforclinicalclassesofobesity.
Consistentwiththeoreticalpredictionsandprevioussmallerstudies,ourresultsshowthatthereisalargeoverlapintermsofgeneticstructureandthedistributionofvariantsbetweentraitsbasedondifferentdistributioncutoffswiththosefrompopulation-levelstudies,butadditionalinsightmaystillbegainedfromevaluat-ingthepopulationextremes.
Ourresultsareinformativefordesigningfuturegeneticstudiesofobesityaswellasothercomplextraits.
MethodsMethodsandanyassociatedreferencesareavailableintheonlineversionofthepaper.
Note:Supplementaryinformationisavailableintheonlineversionofthepaper.
AcknowledgmentsAfulllistofacknowledgmentsappearsintheSupplementaryNote.
FundingwasprovidedbytheAarnoKoskeloFoundation;theAcademyofFinland;theAgencyforScience,TechnologyandResearchofSingapore;theAustralianNationalHealthandMedicalResearchCouncil;theAustralianResearchCouncil;BDAResearch;theBioSHaREConsortium;theBritishHeartFoundation;theCedars-SinaiBoardofGovernors'ChairinMedicalGenetics;theCentreforClinicalResearchattheUniversityofLeipzig;theCentreofExcellenceinGenomicsandtheUniversityofTartu;theChiefScientistOfficeoftheScottishgovernment;theCityofKuopioandtheSocialInsuranceInstitutionofFinland;theDepartmentofEducationalAssistance,theUniversityandResearchoftheAutonomousProvinceofBolzano;theDonaldW.
ReynoldsFoundation;theDutchMinistryforHealth,WelfareandSports;theDutchMinistryofEducation,CultureandScience;DutchBBRMI-NL;theDutchBrainFoundation;theDutchCentreforMedicalSystemsBiology;theDutchDiabetesResearchFoundation;theDutchGovernmentEconomicStructure–EnhancingFund;theDutchInter-UniversityCardiologyInstitute;theDutchKidneyFoundation;theDutchMinistryofEconomicAffairs;theDutchMinistryofJustice;theDutchResearchInstituteforDiseasesintheElderly;EleanorNicholsendowments;theEmilAaltonenFoundation;ErasmusMedicalCenterandErasmusUniversity;theEstoniangovernment;theEuropeanCommission;theEuropeanRegionalDevelopmentFund;theEuropeanResearchCouncil;theEuropeanScienceFoundation;theFacultyofBiologyandMedicineofLausanne;Finland'sSlotMachineAssociation;theFinnishCulturalFoundation;theFinnishDiabetesResearchFoundation;theFinnishFoundationforCardiovascularResearch;theFinnishFundingAgencyforTechnologyandInnovation;theFinnishHeartAssociation;theFinnishMedicalSociety;theFinnishMinistryofEducationandCulture;theFinnishMinistryofHealthandSocialAffairs;theFinnishNationalInstituteforHealthandWelfare;theFinnishSocialInsuranceInstitution;FinskaLkaresllskapet;theFolkhlsanResearchFoundation;theFoundationforLifeandHealthinFinland;theFrenchMinistryofResearch;theFrenchNationalResearchAgency;theGeneticAssociationInformationNetwork;theGermanDiabetesAssociation;theGermanFederalMinistryofEducationandResearch;theGermanMinistryofCulturalAffairs;theGermanNationalGenomeResearchNetwork;theGermanResearchFoundation;GlaxoSmithKline;theGteborgMedicalSociety;theGreekGeneralSecretaryofResearchandTechnology;theGyllenbergFoundation;HealthCareCentersinVasa,NrpesandKorsholm;theHeinzNixdorfFoundation;HelmholtzZentrumMünchen–GermanResearchCenterforEnvironmentalHealth;theIcelandicHeartAssociation;theIcelandicParliament;theIntramuralResearchProgramoftheDivisionofCancerEpidemiologyandGenetics,theNationalCancerInstitute,NIH;ItalianMinistryofEducation,UniversitiesandResearch;ItalianMinistryofHealth;JuhoVainioFoundation;JuvenileDiabetesResearchFoundationInternational;theKnutandAliceWallenbergFoundation;Kuopio,TampereandTurkuUniversityHospitalMedicalFunds;theLeducqFoundation;theLundbergFoundation;theMarchofDimes;theMunichCenterofHealthSciencesaspartofLMUinnovativ;theMunicipalHealthCareCenterandHospitalinJakobstad;theMunicipalityofRotterdam;theNrpesHealthCareFoundation;NationalAllianceforResearchonSchizophreniaandDepressionYoungInvestigatorAwards;theNetherlandsGenomicsInitiative;theNetherlandsOrganizationforHealthResearchandDevelopment;UKNHSBT;USNationalInstitutesofHealth;theNordicCenterofCardiovascularResearch;theNordicCenterofExcellenceinDiseaseGenetics;theNordicCentreofExcellenceonSystemsBiologyinControlledDietaryInterventionsandCohortStudies;theNorthernNetherlandsCollaborationofProvinces;theNovoNordiskFoundation;theOllqvistFoundation;theOrion-FarmosResearchFoundation;thePaavoNurmiFoundation;thePivikkiandSakariSohlbergFoundation;thePerklenFoundation;thePetrusandAugustaHedlundsFoundation;theProvinceofGroningen;theRepublicofCroatiaMinistryofScience,EducationandSport;theReynold'sFoundation;theRoyalSociety;SamfundetFolkhlsan;theSigneandAneGyllenbergFoundation;theSigridJuseliusFoundation;theSocialMinistryoftheFederalStateofMecklenburg–WestPomerania;theSophiaFoundationforMedicalResearch;theSouthTyroleanSparkasseFoundation;theSouthernCaliforniaDiabetesEndocrinologyResearchCenter;theStockholmCountyCouncil;theStrategicCardiovascularProgramofKarolinskaInstitutet;StrategicSupportforEpidemiologicalResearchatKarolinskaInstitutet;theSusanG.
KomenBreastCancerFoundation;theSwedishMinistryforHigherEducation;theSwedishCancerSociety;theSwedishCulturalFoundationinFinland;theSwedishDiabetesAssociation;theSwedishFoundationforStrategicResearch;theSwedishHeart-LungFoundation;theSwedishMedicalResearchCouncil;theSwedishMinistryofNatureGeneticsVOLUME45|NUMBER5|MAY2013507ArticlesEducation;theSwedishResearchCouncil;theSwedishRoyalAcademyofScience;theSwedishSocietyforMedicalResearch;theSwedishSocietyofMedicine;theSwissNationalScienceFoundation;theTampereTuberculosisFoundation;TheGreatWineEstatesoftheMargaretRiverRegionofWesternAustralia;ThePaulMichaelDonovanCharitableFoundation;theTorstenandRagnarSderbergFoundation;CancerResearchUK;theUKDiabetesAssociation;theUKHeartFoundation;theUKMRC;theUKNIHR,BiomedicalResearchCentre;UKWestAngliaPrimaryandCommunityCare;theUniversityMedicalCenterGroningenandtheUniversityofGroningen;theVstraGtalandFoundation;VUUniversity:theInstituteforHealthandCareResearchandtheNeuroscienceCampusAmsterdam;theWellcomeTrust;andtheYrjJahnssonFoundation.
AUTHORCONTRIBUTIONSSteeringcommittee(oversawtheconsortium):G.
R.
A.
,T.
L.
A.
,I.
B.
,S.
I.
B.
,M.
Boehnke,I.
B.
B.
,P.
D.
,C.
S.
F.
,T.
Frayling,L.
C.
G.
,T.
H.
,I.
M.
H.
,D.
H.
,E.
I.
,R.
C.
K.
,R.
J.
F.
L.
,M.
I.
M.
,K.
L.
Mohlke,K.
E.
N.
,J.
R.
O.
,D.
Schlessinger,D.
P.
S.
,U.
T.
andC.
M.
v.
D.
Writinggroup(draftedandeditedthemanuscript):S.
I.
B.
,M.
F.
F.
,A.
Ganna,S.
G.
,E.
I.
,A.
E.
J.
,C.
M.
L.
,R.
J.
F.
L.
,R.
M.
,M.
I.
M.
,D.
Meyre,K.
L.
Monda,A.
P.
M.
,K.
E.
N.
,A.
Scherag,E.
K.
S.
,E.
WheelerandC.
J.
W.
Datacleaningandpreparation:S.
I.
B.
,D.
C.
C.
-C.
,F.
R.
D.
,T.
E.
,T.
Fall,T.
Ferreira,S.
G.
,I.
M.
H.
,E.
I.
,A.
U.
J.
,C.
M.
L.
,J.
Luan,R.
M.
,J.
C.
R.
,A.
Scherag,E.
K.
S.
,G.
T.
,S.
V.
,T.
W.
W.
andA.
R.
W.
Statisticaladvisors:S.
H.
L.
,B.
M.
N.
,Y.
P.
,P.
M.
V.
,J.
Y.
,D.
-Y.
L.
andY.
-J.
H.
Geneexpression(eQTL)analyses:L.
Liang,W.
O.
C.
,M.
F.
M.
,G.
R.
A.
,V.
Steinthorsdottir,G.
T.
,J.
L.
M.
,G.
Nicholson,F.
Karpe.
,M.
I.
M.
andE.
E.
S.
Projectdesign,managementandcoordinationofcontributingstudiesStage1—GWAS:(ADVANCE)T.
L.
A.
andC.
I.
;(AGES)V.
G.
,T.
B.
H.
andL.
J.
L.
;(ARIC)E.
B.
andK.
E.
N.
;(B58C)D.
P.
S.
;(BRIGHTStudy)M.
J.
C.
andP.
B.
M.
;(CAPS)E.
I.
;(CHS)B.
M.
andB.
M.
P.
;(CoLaus)V.
M.
,P.
V.
andG.
Waeber;(COROGENE)M.
S.
N.
andJ.
Sinisalo;(deCODE)K.
StefanssonandU.
T.
;(DGI)L.
C.
G.
andJ.
N.
H.
;(EGCUT)A.
M.
;(EPIC)K.
-T.
K.
,R.
J.
F.
L.
andN.
J.
W.
;(ERF)B.
A.
O.
andC.
M.
v.
D.
;(FamHS)I.
B.
B.
andM.
A.
P.
;(Fenland)R.
J.
F.
L.
andN.
J.
W.
;(FRAM)L.
A.
C.
andC.
S.
F.
;(FUSIONGWAS)M.
BoehnkeandK.
L.
Mohlke;(Genmets)A.
J.
,S.
R.
andV.
Salomaa;(GerMIFS1)J.
E.
andH.
Schunkert;(GerMIFS2)C.
HengstenbergandK.
Stark;(GOOD)C.
O.
;(HBCS)J.
G.
E.
;(KORAS3)H.
-E.
W.
;(KORAS4)C.
G.
,T.
I.
,W.
K.
andA.
Peters;(MGS)P.
V.
G.
andD.
F.
L.
;(MICROS(SOUTHTYROL))P.
P.
P.
;(MIGEN)J.
N.
H.
andS.
Kathiresan;(NESDA)B.
P.
;(NFBC1966)M.
-R.
J.
;(NHS)L.
Q.
;(NijmegenBiomedicalStudy)L.
A.
K.
;(NSPHS)U.
G.
;(NTR)D.
I.
B.
;(ORCADES)J.
F.
W.
andA.
F.
W.
;(PLCO)S.
J.
C.
andS.
I.
B.
;(PROCARDIS)M.
F.
andH.
Watkins;(RS-I)F.
R.
andA.
G.
U.
;(RUNMC)L.
A.
K.
;(SardiNIA)D.
Schlessinger;(SASBAC)E.
I.
;(SHIP)H.
Wallaschofski;(Sorbs)M.
S.
andA.
Tnjes;(TwinsUK)T.
D.
S.
;(VIS)I.
R.
;(WGHS)P.
M.
R.
;(WTCC-T2D)M.
I.
M.
;(WTCCC-CAD)A.
J.
B.
,A.
S.
H.
andN.
J.
S.
;and(YFS)M.
Khnen,T.
L.
,O.
R.
andJ.
Viikari.
Stage2—Metabochipandinsilicoreplication:(AMC-PAS)K.
G.
H.
;(B58C)C.
Power;(BHS)L.
J.
P.
;(DILGOM)K.
KuulasmaaandV.
Salomaa;(DPS)M.
U.
;(DR'sEXTRA)T.
A.
L.
andR.
R.
;(EPIC,FenlandandEly)C.
Langenberg,R.
J.
F.
L.
andN.
J.
W.
;(FIN-D2D2007)S.
M.
K.
-K.
andT.
E.
S.
;(GLACIER)P.
W.
F.
;(Go-DARTS(Dundee))A.
D.
M.
andC.
N.
A.
P.
;(HNR)K.
-H.
J.
;(HUNT2)K.
H.
;(Hypergenes)D.
C.
;(IMPROVE)U.
d.
F.
,A.
HamstenandE.
T.
;(KORAS3)I.
M.
H.
;(LifeLinesCohortStudy)H.
Snieder,M.
M.
V.
d.
K.
andB.
H.
R.
W.
;(LURIC)B.
O.
B.
,W.
M.
andB.
R.
W.
;(METSIM)J.
K.
andM.
Laakso;(MORGAM)P.
A.
,P.
B.
,M.
M.
F.
,J.
F.
,F.
Kee,D.
-A.
T.
andJ.
Virtamo;(NSHD)D.
K.
;(PIVUS)E.
I.
;(PLCO2)S.
I.
B.
andS.
J.
C.
;(PREVEND)P.
v.
d.
H.
;(QIMR)N.
G.
M.
,G.
W.
M.
,A.
C.
H.
andP.
M.
;(RS-II)A.
HofmanandJ.
B.
J.
v.
M.
;(RS-III)C.
M.
v.
D.
andJ.
C.
M.
W.
;(SwedishTwinRegistry)E.
I.
;(THISEAS/AMCPAS/CARDIOGENICS)P.
D.
;(THISEAS)G.
V.
D.
;(TRAILS)A.
J.
O.
;(Troms4)I.
N.
;(TWINGENE)E.
I.
;(UKBS2)W.
H.
O.
;(ULSAM)E.
I.
;(WhitehallII)A.
HingoraniandM.
Kivimki;and(WTCC-T2D)M.
I.
M.
andC.
M.
L.
Othercontributingstudies,clinicalextremes:(FrenchExtremeObesityStudy)D.
MeyreandP.
F.
;(GEO-IT)A.
M.
D.
B.
;(EssenObesityStudy,EssenCase-Control&EssenObesityTrioGWAS)J.
H.
andA.
Hinney;and(GOYA)T.
I.
A.
S.
andE.
A.
N.
GenotypingofcontributingstudiesStage1—GWAS:(ADVANCE)D.
A.
;(ARIC)E.
B.
;(B58C)W.
L.
M.
;(CAPS)H.
Grnberg;(CHS)T.
H.
;(CoLaus)V.
M.
;(COROGENE)M.
Perola;(EGCUT)T.
E.
andL.
M.
;(EPIC)I.
B.
;(ERF)B.
A.
O.
andC.
M.
v.
D.
;(FamHS)I.
B.
B.
,M.
A.
P.
andA.
T.
K.
;(Fenland)J.
Luan;(Genmets)S.
R.
;(GOOD)C.
O.
,J.
-O.
J.
andM.
Lorentzon;(HBCS)A.
PalotieandE.
Widén;(MGS)P.
V.
G.
andA.
R.
S.
;(MICROS(SOUTHTYROL))A.
A.
H.
;(NHS)F.
B.
H.
andD.
H.
;(NSPHS).
J.
;(NTRandNESDA)J.
-J.
H.
;(ORCADES)J.
F.
W.
;(PLCO)S.
J.
C.
andK.
B.
J.
;(RS-I)F.
R.
,A.
G.
U.
,K.
E.
andC.
M.
-G.
;(SardiNIA)M.
Dei;(SASBAC)P.
H.
andJ.
Liu;(SHIP)G.
H.
;(TwinsUK)M.
M.
,S.
-Y.
S.
andN.
S.
;(VIS)C.
HaywardandV.
V.
;(WGHS)D.
I.
C.
;(WTCC-T2D)M.
I.
M.
;(WTCCC-CAD)A.
J.
B.
,A.
S.
H.
andN.
J.
S.
;and(YFS)T.
L.
Stage2—Metabochipandinsilicoreplication:(BHS)L.
J.
P.
andJ.
B.
;(CARDIOGENICS)S.
E.
andS.
E.
H.
;(DPS)A.
J.
S.
;(DR'sEXTRA)M.
A.
;(EPIC,FenlandandEly)J.
LuanandK.
K.
O.
;(FIN-D2D2007)P.
S.
C.
;(FUSION)F.
S.
C.
,J.
SaramiesandJ.
Tuomilehto;(GLACIER)I.
B.
andS.
E.
;(Go-DARTS(Dundee))C.
N.
A.
P.
;(HNR)T.
W.
M.
;(HUNT2)N.
N.
;(Hypergenes)F.
F.
;(KORAS3)H.
Grallert;(KORAS4)T.
I.
;(LifeLinesCohortStudy)H.
Snieder,B.
H.
R.
W.
,M.
BruinenbergandL.
F.
;(NSHD)D.
K.
,K.
K.
O.
andA.
W.
;(PIVUS)E.
I.
andL.
Lind;(PLCO2)S.
J.
C.
,K.
B.
J.
andZ.
W.
;(PREVEND)P.
v.
d.
H.
andF.
W.
A.
;(QIMR)N.
G.
M.
,G.
W.
M.
,A.
C.
H.
andP.
A.
M.
;(RS-II)M.
J.
P.
andM.
Dei;(SwedishTwinRegistry)E.
I.
,P.
K.
M.
andN.
P.
;(THISEAS/AMCPAS/CARDIOGENICS)K.
Stirrups;(TRAILS)A.
J.
O.
,I.
M.
N.
andJ.
V.
V.
V.
-O.
;(Troms4)L.
L.
B.
;(TWINGENE)E.
I.
,A.
HamstenandN.
P.
;(ULSAM)E.
I.
;(WhitehallII)C.
Langenberg;and(WTCC-T2D)M.
I.
M.
Othercontributingstudies,clinicalextremes:(FrenchExtremeObesityStudy)D.
MeyreandP.
F.
;(GEO-IT)D.
G.
;and(GOYA)L.
P.
andD.
M.
E.
PhenotypingofcontributingstudiesStage1—GWAS:(ARIC)E.
B.
;(B58C)D.
P.
S.
;(BRIGHTStudy)J.
M.
C.
;(CAPS)H.
Grnberg;(CHS)B.
M.
P.
;(CoLaus)P.
V.
andG.
Waeber;(COROGENE)J.
Sinisalo,M.
-L.
L.
;(EGCUT)A.
M.
andK.
F.
;(EPIC)R.
J.
F.
L.
;(ERF)B.
A.
O.
andC.
M.
v.
D.
;(FamHS)I.
B.
B.
,M.
A.
P.
andM.
F.
F.
;(Fenland)R.
J.
F.
L.
;(FRAM)C.
S.
F.
;(Genmets)A.
J.
andV.
Salomaa;(GerMIFS1)S.
Schreiber;(GerMIFS2)A.
Peters;(GOOD)C.
O.
,J.
-O.
J.
,M.
LorentzonandL.
V.
;(MGS)P.
V.
G.
,A.
R.
S.
andD.
F.
L.
;(NFBC1966)A.
-L.
H.
,J.
H.
L.
andA.
Pouta;(NHS)L.
Q.
;(NijmegenBiomedicalStudy)F.
d.
V.
,M.
d.
H.
andS.
H.
V.
;(NSPHS).
J.
andU.
G.
;(NTRandNESDA)G.
Willemsen;(ORCADES)H.
C.
andS.
H.
W.
;(PLCO)S.
I.
B.
;(RS-I)F.
R.
andA.
G.
U.
;(SASBAC)P.
H.
;(SHIP)S.
Schipf;(Sorbs)A.
Tnjes;(TwinsUK)M.
M.
andT.
D.
S.
;(VIS)O.
P.
;(WTCC-T2D)M.
I.
M.
;(WTCCC-CAD)A.
J.
B.
,A.
S.
H.
andN.
J.
S.
;and(YFS)M.
Khnen,O.
R.
andJ.
Viikari.
Stage2—Metabochipandinsilicoreplication:(AMC-PAS)H.
B.
andM.
D.
T.
;(B58C)C.
PowerandE.
H.
;(BHS)L.
J.
P.
,J.
B.
andA.
W.
M.
;(DPS)J.
Lindstrm;(EPIC,FenlandandEly)R.
J.
F.
L.
;(GLACIER)P.
W.
F.
andD.
Shungin;(Go-DARTS(Dundee))C.
N.
A.
P.
andA.
D.
M.
;(Hypergenes)D.
C.
andP.
M.
;(KORAS3)B.
T.
;(KORAS4)A.
Peters;(LifeLinesCohortStudy)B.
H.
R.
W.
andM.
M.
V.
d.
K.
;(METSIM)A.
StanákováandP.
V.
G.
;(NSHD)D.
K.
;(PIVUS)E.
I.
andL.
Lind;(PLCO2)S.
I.
B.
;(PREVEND)G.
Navis;(QIMR)N.
G.
M.
,A.
C.
H.
andP.
M.
;(RS-II)M.
C.
Z.
;(RS-III)J.
C.
M.
W.
;(SwedishTwinRegistry)E.
I.
,P.
K.
M.
andN.
P.
;(THISEAS)M.
DimitriouandE.
V.
T.
;(TRAILS)R.
P.
S.
;(TWINGENE)E.
I.
andN.
P.
;(UKBS2)A.
Rendon;(ULSAM)E.
I.
;(WhitehallII)M.
Kumari;and(WTCC-T2D)M.
I.
M.
Othercontributingstudies,clinicalextremes:(EssenObesityStudy,EssenCase-ControlGWAS&EssenObesityTrioGWAS)J.
H.
andA.
Hinney;(GEO-IT)A.
L.
andS.
Signorini;and(GOYA)T.
I.
A.
S.
andE.
A.
N.
AnalysesofcontributingstudiesStage1—GWAS:(ADVANCE)L.
L.
W.
;(AGES)A.
V.
S.
;(ARIC)K.
E.
N.
,A.
E.
J.
andK.
L.
Monda;(B58C)D.
P.
S.
;(BRIGHTStudy)T.
J.
;(CAPS)E.
I.
andR.
M.
;(CHS)B.
M.
andG.
L.
;(CoLaus)D.
Marek;(COROGENE)M.
Perola;(deCODE)V.
SteinthorsdottirandG.
T.
;(DGI)E.
K.
S.
andS.
V.
;(EGCUT)K.
F.
,T.
E.
andE.
M.
;(EPIC)J.
H.
Z.
;(ERF)N.
A.
;(FamHS)M.
F.
F.
;(Fenland)J.
Luan;(FRAM)L.
A.
C.
,N.
L.
H.
-C.
andJ.
S.
N.
;(Genmets)I.
S.
;(GerMIFS1)M.
Preuss;(GerMIFS2)I.
R.
K.
;(GOOD)C.
O.
,J.
-O.
J.
,M.
LorentzonandL.
V.
;(HBCS)N.
E.
;(KORAS3)C.
Lamina;(KORAS4)E.
A.
;(MGS)D.
F.
L.
andJ.
Shi;(MICROS(SOUTHTYROL))J.
E.
H.
and.
J.
;(MIGEN)E.
K.
S.
andS.
V.
;(NFBC1966)A.
Pouta,R.
M.
andJ.
C.
R.
;(NHS)L.
Q.
andT.
W.
;(NSPHS).
J.
;(NTRandNESDA)J.
-J.
H.
;(ORCADES).
J.
;(PLCO)K.
B.
J.
andS.
I.
B.
;(PROCARDIS)M.
F.
,A.
GoelandJ.
F.
P.
;(RS-I)F.
R.
,K.
E.
andC.
M.
-G.
;(SardiNIA)J.
L.
B.
-G.
andS.
Sanna;(SASBAC)E.
I.
andR.
M.
;(SEARCH)J.
Tyrer;(SHIP)A.
Teumer;(Sorbs)R.
M.
andI.
P.
;(TwinsUK)M.
M.
andN.
S.
;(VIS).
J.
;(WGHS)D.
I.
C.
;(WTCC-T2D)C.
M.
L.
,R.
M.
andJ.
C.
R.
;(WTCCC-CAD)R.
M.
andJ.
C.
R.
;(WTCCC-NBS(UKBS-CC))A.
P.
A.
,R.
M.
,J.
C.
R.
,J.
G.
S.
andJ.
C.
S.
;and(YFS)O.
R.
andT.
L.
Stage2—Metabochipandinsilicoreplication:(B58C)E.
H.
andT.
Ferreira;(BHS)G.
C.
;(DILGOM)K.
KristianssonandK.
Kuulasmaa;(DPS)A.
U.
J.
;(DR'sEXTRA)A.
U.
J.
;(EPIC,FenlandandEly)J.
LuanandK.
K.
O.
;(FIN-D2D2007)A.
U.
J.
;(FUSION)A.
U.
J.
;(GLACIER)P.
W.
F.
andD.
Shungin;(HNR)S.
P.
andC.
Pütter;(HUNT2)A.
U.
J.
;(Hypergenes)F.
F.
andZ.
K.
;(IMPROVE)R.
J.
S.
;(KORAS3)I.
M.
H.
andT.
W.
W.
;(KORAS4)M.
M.
-N.
;(LifeLinesCohortStudy)M.
BruinenbergandL.
F.
;(LURIC)M.
E.
K.
;(METSIM)A.
U.
J.
;(NSHD)A.
W.
andJ.
Luan;(PIVUS)E.
I.
,S.
G.
;(PLCO2)S.
I.
B.
,Z.
W.
;(PREVEND)P.
v.
d.
H.
andI.
M.
L.
;(QIMR)S.
E.
M.
,J.
Y.
;(RS-II)M.
J.
P.
;(SwedishTwinRegistry)E.
I.
andS.
G.
;(THISEAS/AMCPAS/CARDIOGENICS)S.
Kanoni;(TRAILS)I.
M.
N.
andJ.
V.
V.
V.
-O.
;(Troms4)A.
U.
J.
;(TWINGENE)E.
I.
andS.
G.
;(UKBS2)A.
Radhakrishnan;(ULSAM)E.
I.
,A.
GannaandS.
G.
;(WGHS)L.
M.
R.
;and(WTCC-T2D)R.
M.
andT.
Ferreira.
Othercontributingstudies,clinicalextremes:(FrenchExtremeObesityStudy)D.
Meyre,C.
LecoeurandB.
S.
;(GEO-IT)A.
M.
D.
B.
andD.
G.
;(EssenObesityStudy,EssenCase-ControlGWASandEssenObesityTrioGWAS)A.
ScheragandI.
J.
;and(GOYA)L.
P.
andD.
M.
E.
COMPETINGFINANCIALINTERESTSTheauthorsdeclarecompetingfinancialinterests:detailsareavailableintheonlineversionofthepaper.
508VOLUME45|NUMBER5|MAY2013NatureGeneticsArticlesReprintsandpermissionsinformationisavailableonlineathttp://www.
nature.
com/reprints/index.
html.
1.
Maes,H.
H.
,Neale,M.
C.
&Eaves,L.
J.
Geneticandenvironmentalfactorsinrelativebodyweightandhumanadiposity.
Behav.
Genet.
27,325–351(1997).
2.
Stunkard,A.
J.
,Foch,T.
T.
&Hrubec,Z.
Atwinstudyofhumanobesity.
J.
Am.
Med.
Assoc.
256,51–54(1986).
3.
Silventoinen,K.
etal.
Heritabilityofadultbodyheight:acomparativestudyoftwincohortsineightcountries.
TwinRes.
6,399–408(2003).
4.
Speliotes,E.
K.
etal.
Associationanalysesof249,796individualsreveal18newlociassociatedwithbodymassindex.
Nat.
Genet.
42,937–948(2010).
5.
Okada,Y.
etal.
CommonvariantsatCDKAL1andKLF9areassociatedwithbodymassindexineastAsianpopulations.
Nat.
Genet.
44,302–306(2012).
6.
Wen,W.
etal.
Meta-analysisidentifiescommonvariantsassociatedwithbodymassindexineastAsians.
Nat.
Genet.
44,307–311(2012).
7.
Heid,I.
M.
etal.
Meta-analysisidentifies13newlociassociatedwithwaist-hipratioandrevealssexualdimorphisminthegeneticbasisoffatdistribution.
Nat.
Genet.
42,949–960(2010).
8.
Lindgren,C.
M.
etal.
Genome-wideassociationscanmeta-analysisidentifiesthreelociinfluencingadiposityandfatdistribution.
PLoSGenet.
5,e1000508(2009).
9.
LangoAllen,H.
etal.
Hundredsofvariantsclusteredingenomiclociandbiologicalpathwaysaffecthumanheight.
Nature467,832–838(2010).
10.
Lee,S.
H.
,Wray,N.
R.
,Goddard,M.
E.
&Visscher,P.
M.
Estimatingmissingheritabilityfordiseasefromgenome-wideassociationstudies.
Am.
J.
Hum.
Genet.
88,294–305(2011).
11.
Zuk,O.
,Hechter,E.
,Sunyaev,S.
R.
&Lander,E.
S.
Themysteryofmissingheritability:geneticinteractionscreatephantomheritability.
Proc.
Natl.
Acad.
Sci.
USA109,1193–1198(2012).
12.
Duncan,E.
L.
etal.
Genome-wideassociationstudyusingextremetruncateselectionidentifiesnovelgenesaffectingbonemineraldensityandfracturerisk.
PLoSGenet.
7,e1001372(2011).
13.
Edmondson,A.
C.
etal.
Densegenotypingofcandidategenelociidentifiesvariantsassociatedwithhigh-densitylipoproteincholesterol.
Circ.
Cardiovasc.
Genet.
4,145–155(2011).
14.
Meyre,D.
etal.
Genome-wideassociationstudyforearly-onsetandmorbidadultobesityidentifiesthreenewrisklociinEuropeanpopulations.
Nat.
Genet.
41,157–159(2009).
15.
Scherag,A.
etal.
Twonewlociforbody-weightregulationidentifiedinajointanalysisofgenome-wideassociationstudiesforearly-onsetextremeobesityinFrenchandGermanstudygroups.
PLoSGenet.
6,e1000916(2010).
16.
Bradfield,J.
P.
etal.
Agenome-wideassociationmeta-analysisidentifiesnewchildhoodobesityloci.
Nat.
Genet.
44,526–531(2012).
17.
Cohen,J.
C.
etal.
MultiplerareallelescontributetolowplasmalevelsofHDLcholesterol.
Science305,869–872(2004).
18.
Emond,M.
J.
etal.
ExomesequencingofextremephenotypesidentifiesDCTN4asamodifierofchronicPseudomonasaeruginosainfectionincysticfibrosis.
Nat.
Genet.
44,886–889(2012).
19.
Harismendy,O.
etal.
Populationsequencingoftwoendocannabinoidmetabolicgenesidentifiesrareandcommonregulatoryvariantsassociatedwithextremeobesityandmetabolitelevel.
GenomeBiol.
11,R118(2010).
20.
Romeo,S.
etal.
Population-basedresequencingofANGPTL4uncoversvariationsthatreducetriglyceridesandincreaseHDL.
Nat.
Genet.
39,513–516(2007).
21.
Chan,Y.
etal.
Commonvariantsshowpredictedpolygeniceffectsonheightinthetailsofthedistribution,exceptinextremelyshortindividuals.
PLoSGenet.
7,e1002439(2011).
22.
Cotsapas,C.
etal.
Commonbodymassindex–associatedvariantsconferriskofextremeobesity.
Hum.
Mol.
Genet.
18,3502–3507(2009).
23.
Paternoster,L.
etal.
Genome-widepopulation-basedassociationstudyofextremelyoverweightyoungadults—theGOYAstudy.
PLoSONE6,e24303(2011).
24.
Hinney,A.
etal.
Genomewideassociation(GWA)studyforearlyonsetextremeobesitysupportstheroleoffatmassandobesityassociatedgene(FTO)variants.
PLoSONE2,e1361(2007).
25.
Benzinou,M.
etal.
CommonnonsynonymousvariantsinPCSK1conferriskofobesity.
Nat.
Genet.
40,943–945(2008).
26.
Jiao,H.
etal.
GenomewideassociationstudyidentifiesKCNMA1contributingtohumanobesity.
BMCMed.
Genomics4,51(2011).
27.
denHoed,M.
etal.
EvaluationofcommongeneticvariantsidentifiedbyGWASforearlyonsetandmorbidobesityinpopulation-basedsamples.
Int.
J.
Obes.
(Lond).
37,191–196(2013).
28.
Willer,C.
J.
etal.
Sixnewlociassociatedwithbodymassindexhighlightaneuronalinfluenceonbodyweightregulation.
Nat.
Genet.
41,25–34(2009).
29.
Pütter,C.
etal.
MissingheritabilityinthetailsofquantitativetraitsAsimulationstudyontheimpactofslightlyalteredtruegeneticmodels.
Hum.
Hered.
72,173–181(2011).
30.
Williams,P.
T.
Quantile-specificpenetranceofgenesaffectinglipoproteins,adiposityandheight.
PLoSONE7,e28764(2012).
31.
Guey,L.
T.
etal.
Powerinthephenotypicextremes:asimulationstudyofpowerindiscoveryandreplicationofrarevariants.
Genet.
Epidemiol.
Publishedonlinedoi:10.
1002/gepi.
20572(9February2011).
32.
WorldHealthOrganization.
Obesity:preventingandmanagingtheglobalepidemic.
ReportofaWHOConsultation.
inWHOTechnicalReportSeries8949(WorldHealthOrganization,Geneva,2000).
33.
Kumanyika,S.
K.
etal.
Population-basedpreventionofobesity:theneedforcomprehensivepromotionofhealthfuleating,physicalactivity,andenergybalance:ascientificstatementfromAmericanHeartAssociationCouncilonEpidemiologyandPrevention,InterdisciplinaryCommitteeforPrevention(formerlytheexpertpanelonpopulationandpreventionscience).
Circulation118,428–464(2008).
34.
Sarbassov,D.
D.
&Sabatini,D.
M.
Redoxregulationofthenutrient-sensitiveraptor-mTORpathwayandcomplex.
J.
Biol.
Chem.
280,39505–39509(2005).
35.
Daigo,K.
etal.
Proteomicanalysisofnativehepatocytenuclearfactor-4α(HNF4α)isoforms,phosphorylationstatus,andinteractivecofactors.
J.
Biol.
Chem.
286,674–686(2011).
36.
Nakajima,H.
etal.
Hepatocytenuclearfactor-4αgenemutationsinJapanesenon-insulindependentdiabetesmellitus(NIDDM)patients.
Res.
Commun.
Mol.
Pathol.
Pharmacol.
94,327–330(1996).
37.
Cho,Y.
S.
etal.
Meta-analysisofgenome-wideassociationstudiesidentifieseightnewlocifortype2diabetesineastAsians.
Nat.
Genet.
44,67–72(2012).
38.
Dupuis,J.
etal.
Newgeneticlociimplicatedinfastingglucosehomeostasisandtheirimpactontype2diabetesrisk.
Nat.
Genet.
42,105–116(2010).
39.
Saxena,R.
etal.
GeneticvariationinGIPRinfluencestheglucoseandinsulinresponsestoanoralglucosechallenge.
Nat.
Genet.
42,142–148(2010).
40.
Shi,J.
&Kandror,K.
V.
SortilinisessentialandsufficientfortheformationofGlut4storagevesiclesin3T3-L1adipocytes.
Dev.
Cell9,99–108(2005).
41.
Kaddai,V.
etal.
InvolvementofTNF-αinabnormaladipocyteandmusclesortilinexpressioninobesemiceandhumans.
Diabetologia52,932–940(2009).
42.
Zhang,M.
etal.
ParacrineoverexpressionofIGFBP-4inosteoblastsoftransgenicmicedecreasesboneturnoverandcausesglobalgrowthretardation.
J.
BoneMiner.
Res.
18,836–843(2003).
43.
Liao,Y.
C.
,Chen,N.
T.
,Shih,Y.
P.
,Dong,Y.
&Lo,S.
H.
Up-regulationofC-terminaltensin-likemoleculepromotesthetumorigenicityofcoloncancerthroughβ-catenin.
CancerRes.
69,4563–4566(2009).
44.
Milat,F.
&Ng,K.
W.
IsWntsignallingthefinalcommonpathwayleadingtoboneformationMol.
CellEndocrinol.
310,52–62(2009).
45.
Purcell,S.
M.
etal.
Commonpolygenicvariationcontributestoriskofschizophreniaandbipolardisorder.
Nature460,748–752(2009).
46.
Yang,J.
etal.
Conditionalandjointmultiple-SNPanalysisofGWASsummarystatisticsidentifiesadditionalvariantsinfluencingcomplextraits.
Nat.
Genet.
44,369–375(2012).
47.
Kilpelinen,T.
O.
,Bingham,S.
A.
,Khaw,K.
T.
,Wareham,N.
J.
&Loos,R.
J.
AssociationofvariantsinthePCSK1genewithobesityintheEPIC-Norfolkstudy.
Hum.
Mol.
Genet.
18,3496–3501(2009).
SonjaIBerndt1,244,StefanGustafsson2,3,244,ReedikMgi4,5,244,AndreaGanna3,244,EleanorWheeler6,MaryFFeitosa7,AnneEJustice8,KeriLMonda8,9,DamienCCroteau-Chonka10,FelixRDay11,TnuEsko5,12,ToveFall3,TeresaFerreira4,DavideGentilini13,AnneUJackson14,Jian'anLuan11,JoshuaCRandall4,6,SailajaVedantam15–17,CristenJWiller18–20,ThomasWWinkler21,AndrewRWood22,TsegaselassieWorkalemahu23,24,Yi-JuanHu25,SangHongLee26,LimingLiang27,28,Dan-YuLin29,JosineLMin4,BenjaminMNeale30,GudmarThorleifsson31,JianYang32,33,EvaAlbrecht34,NajafAmin35,JenniferLBragg-Gresham14,GemmaCadby36–38,MartindenHeijer39,NiinaEklund40,KristaFischer5,AnujGoel41,Jouke-JanHottenga42,JenniferEHuffman43,IvonneJarick44,saJohansson45,46,TobyJohnson47,48,StavroulaKanoni6,MarcusEKleber49,50,InkeRKnig51,KatiKristiansson40,ZoltánKutalik52,53,ClaudiaLamina54,CecileLecoeur55,56,GuoLi57,MassimoMangino58,WendyLMcArdle59,NatureGeneticsVOLUME45|NUMBER5|MAY2013509ArticlesCarolinaMedina-Gomez35,60,61,MartinaMüller-Nurasyid34,62–64,JuliusSNgwa65,IljaMNolte66,LaviniaPaternoster67,SonaliPechlivanis68,MarkusPerola5,40,69,MarjoleinJPeters35,60,61,MichaelPreuss51,70,LyndaMRose71,JianxinShi1,DmitryShungin72–74,AlbertVernonSmith75,76,RonaJStrawbridge77,IdaSurakka40,69,AlexanderTeumer78,MiekeDTrip79,80,JonathanTyrer81,JanaVVanVliet-Ostaptchouk82,83,LiesbethVandenput84,LindsayLWaite85,JingHuaZhao11,DevinAbsher85,FolkertWAsselbergs86,MustafaAtalay87,AntonyPAttwood88,AnthonyJBalmforth89,HannekeBasart79,JohnBeilby90,91,LoriLBonnycastle92,PaoloBrambilla93,MarcelBruinenberg83,HarryCampbell94,DanielIChasman71,95,PeterSChines92,FrancisSCollins92,JohnMConnell96,97,WilliamOCookson98,UlfdeFaire99,FemmiedeVegt100,MarianoDei101,MariaDimitriou102,SarahEdkins6,KarolEstrada35,60,61,DavidMEvans67,MartinFarrall41,MarcoMFerrario103,JeanFerrières104,LudeFranke83,105,FrancescaFrau106,PabloVGejman107,108,HaraldGrallert109,HenrikGrnberg3,VilmundurGudnason75,76,AlistairSHall110,PerHall3,Anna-LiisaHartikainen111,CarolineHayward43,NancyLHeard-Costa112,AndrewCHeath113,JohannesHebebrand114,GeorgHomuth78,FrankBHu23,SarahEHunt6,ElinaHyppnen115,CarlosIribarren116,KevinBJacobs1,117,John-OlovJansson118,AnttiJula119,MikaKhnen120,SekarKathiresan16,121–123,FrankKee124,Kay-TeeKhaw125,MikaKivimki126,WolfgangKoenig127,AldiTKraja7,MeenaKumari126,KariKuulasmaa128,JohannaKuusisto129,JaanaHLaitinen130,TimoALakka87,131,ClaudiaLangenberg11,126,LenoreJLauner132,LarsLind133,JaanaLindstrm134,JianjunLiu135,AntonioLiuzzi136,Marja-LiisaLokki137,MattiasLorentzon84,PamelaAMadden113,PatrikKMagnusson3,PaoloManunta138,DianaMarek52,53,WinfriedMrz50,139,IreneMateoLeach140,BarbaraMcKnight141,SarahEMedland33,EvelinMihailov5,12,LiliMilani5,GrantWMontgomery33,VincentMooser142,ThomasWMühleisen143,144,PatriciaBMunroe47,48,ArthurWMusk145–147,NarisuNarisu92,GerjanNavis148,GeorgeNicholson149,150,EllenANohr151,KenKOng11,152,BenAOostra61,153,154,ColinNAPalmer155,AarnoPalotie6,69,JohnFPeden156,NancyPedersen3,AnnettePeters109,157,158,OzrenPolasek159,AnneliPouta111,160,PeterPPramstaller161–163,IngaProkopenko4,164,CarolinPütter68,AparnaRadhakrishnan6,165,166,OlliRaitakari167,168,AugustoRendon88,165,166,169,FernandoRivadeneira35,60,61,IgorRudan94,TimoESaaristo170,171,JenniferGSambrook165,166,AlanRSanders107,108,SerenaSanna101,JoukoSaramies172,SabineSchipf173,StefanSchreiber174,HeribertSchunkert175,So-YounShin6,StefanoSignorini176,JuhaSinisalo177,BorisSkrobek55,56,NicoleSoranzo6,58,AlenaStanáková178,KlausStark179,JonathanCStephens165,166,KathleenStirrups6,RonaldPStolk66,83,MichaelStumvoll180,181,AmyJSwift92,EiriniVTheodoraki102,BarbaraThorand157,David-AlexandreTregouet182,ElenaTremoli183,MelanieMVanderKlauw82,83,JoyceBJvanMeurs35,60,61,SitaHVermeulen100,184,JormaViikari185,JarmoVirtamo128,VeroniqueVitart43,GérardWaeber186,ZhaomingWang1,117,ElisabethWidén69,SarahHWild94,GonnekeWillemsen42,BernhardRWinkelmann187,JacquelineCMWitteman35,61,BruceHRWolffenbuttel82,83,AndrewWong152,AlanFWright43,MCarolaZillikens60,61,PhilippeAmouyel188,BernhardOBoehm189,EricBoerwinkle190,191,DorretIBoomsma42,MarkJCaulfield47,48,StephenJChanock1,LAdrienneCupples65,DanieleCusi106,192,GeorgeVDedoussis102,JeanetteErdmann70,175,JohanGEriksson193–195,PaulWFranks23,72,73,PhilippeFroguel55,56,196,ChristianGieger34,UlfGyllensten45,AndersHamsten77,TamaraBHarris132,ChristianHengstenberg179,AndrewAHicks161,AroonHingorani126,AnkeHinney114,AlbertHofman35,61,KeesGHovingh79,KristianHveem197,ThomasIllig109,198,Marjo-RiittaJarvelin160,199–201,Karl-HeinzJckel68,SirkkaMKeinanen-Kiukaanniemi201,202,LambertusAKiemeney100,203,204,DianaKuh152,MarkkuLaakso129,TerhoLehtimki205,DouglasFLevinson206,NicholasGMartin33,AndresMetspalu5,12,AndrewDMorris155,MarkkuSNieminen177,IngerNjlstad207,208,ClaesOhlsson85,AlbertineJOldehinkel209,WillemHOuwehand6,88,165,166,LyleJPalmer36,37,BrendaPenninx210,ChrisPower115,MichaelAProvince7,BruceMPsaty57,211–214,LuQi23,24,RainerRauramaa131,215,PaulMRidker71,95,SamuliRipatti6,41,69,VeikkoSalomaa128,NileshJSamani216,217,HaroldSnieder66,83,ThorkildIASrensen218,219,TimothyDSpector58,KariStefansson31,76,AnkeTnjes180,181,JaakkoTuomilehto134,220,221,AndréGUitterlinden35,60,61,MattiUusitupa222,223,PimvanderHarst105,140,PeterVollenweider186,HenriWallaschofski224,NicholasJWareham11,HughWatkins41,H-ErichWichmann63,225,226,JamesFWilson94,GoncaloRAbecasis14,ThemistoclesLAssimes227,InêsBarroso6,228,MichaelBoehnke14,IngridBBorecki7,PanosDeloukas6,CarolineSFox229,TimothyFrayling22,LeifCGroop230,TalinHaritunian231,IrisMHeid21,225,DavidHunter23,24,RobertCKaplan232,FredrikKarpe164,233,MiriamFMoffatt98,KarenLMohlke10,JeffreyRO'Connell234,510VOLUME45|NUMBER5|MAY2013NatureGeneticsArticlesYudiPawitan3,EricESchadt235,236,DavidSchlessinger237,ValgerdurSteinthorsdottir31,DavidPStrachan238,UnnurThorsteinsdottir31,76,CorneliaMvanDuijn35,61,154,PeterMVisscher26,32,AnnaMariaDiBlasio13,JoelNHirschhorn15–17,CeciliaMLindgren4,AndrewPMorris4,DavidMeyre55,56,239,AndréScherag68,MarkIMcCarthy4,164,233,245,ElizabethKSpeliotes20,240,245,KariENorth8,245,RuthJFLoos11,241–243,245&ErikIngelsson2–4,221USDepartmentofHealthandHumanServices,DivisionofCancerEpidemiologyandGenetics,NationalCancerInstitute,USNationalInstitutesofHealth,Bethesda,Maryland,USA.
2DepartmentofMedicalSciences,MolecularEpidemiologyandScienceforLifeLaboratory,UppsalaUniversity,Uppsala,Sweden.
3DepartmentofMedicalEpidemiologyandBiostatistics,KarolinskaInstitutet,Stockholm,Sweden.
4WellcomeTrustCentreforHumanGenetics,UniversityofOxford,Oxford,UK.
5EstonianGenomeCenter,UniversityofTartu,Tartu,Estonia.
6WellcomeTrustSangerInstitute,Hinxton,Cambridge,UK.
7DepartmentofGenetics,WashingtonUniversitySchoolofMedicine,St.
Louis,Missouri,USA.
8DepartmentofEpidemiology,SchoolofPublicHealth,UniversityofNorthCarolinaatChapelHill,ChapelHill,NorthCarolina,USA.
9CenterforObservationalResearch,Amgen,ThousandsOaks,California,USA.
10DepartmentofGenetics,UniversityofNorthCarolinaatChapelHill,ChapelHill,NorthCarolina,USA.
11MedicalResearchCouncil(MRC)EpidemiologyUnit,InstituteofMetabolicScience,Addenbrooke'sHospital,Cambridge,UK.
12InstituteofMolecularandCellBiology,UniversityofTartu,Tartu,Estonia.
13MolecularBiologyDepartment,IstitutoAuxologicoItaliano,Milan,Italy.
14DepartmentofBiostatistics,CenterforStatisticalGenetics,UniversityofMichigan,AnnArbor,Michigan,USA.
15DivisionsofGeneticsandEndocrinologyandCenterforBasicandTranslationalObesityResearch,Children'sHospital,Boston,Massachusetts,USA.
16MetabolismInitiativeandPrograminMedicalandPopulationGenetics,BroadInstitute,Cambridge,Massachusetts,USA.
17DepartmentofGenetics,HarvardMedicalSchool,Boston,Massachusetts,USA.
18DepartmentofInternalMedicine(Cardiovascular),UniversityofMichigan,AnnArbor,Michigan,USA.
19DepartmentofHumanGenetics,UniversityofMichigan,AnnArbor,Michigan,USA.
20DepartmentofComputationalMedicineandBioinformatics,UniversityofMichigan,AnnArbor,Michigan,USA.
21PublicHealthandGenderStudies,InstituteofEpidemiologyandPreventiveMedicine,RegensburgUniversityMedicalCenter,Regensburg,Germany.
22GeneticsofComplexTraits,PeninsulaCollegeofMedicineandDentistry,UniversityofExeter,Exeter,UK.
23DepartmentofNutrition,HarvardSchoolofPublicHealth,Boston,Massachusetts,USA.
24ChanningLaboratory,DepartmentofMedicine,BrighamandWomen'sHospitalandHarvardMedicalSchool,Boston,Massachusetts,USA.
25DepartmentofBiostatisticsandBioinformatics,EmoryUniversity,Atlanta,Georgia,USA.
26TheQueenslandBrainInstitute,TheUniversityofQueensland,Brisbane,Queensland,Australia.
27DepartmentofEpidemiology,HarvardSchoolofPublicHealth,Boston,Massachusetts,USA.
28DepartmentofBiostatistics,HarvardSchoolofPublicHealth,Boston,Massachusetts,USA.
29DepartmentofBiostatistics,UniversityofNorthCarolinaatChapelHill,ChapelHill,NorthCarolina,USA.
30AnalyticandTranslationalGeneticsUnit,MassachusettsGeneralHospital,Boston,Massachusetts,USA.
31deCODEGenetics,Reykjavik,Iceland.
32UniversityofQueenslandDiamantinaInstitute,TheUniversityofQueensland,PrincessAlexandraHospital,Brisbane,Queensland,Australia.
33QueenslandInstituteofMedicalResearch,Brisbane,Queensland,Australia.
34InstituteofGeneticEpidemiology,HelmholtzZentrumMünchen–GermanResearchCenterforEnvironmentalHealth,Neuherberg,Germany.
35DepartmentofEpidemiology,ErasmusMedicalCenter,Rotterdam,TheNetherlands.
36GeneticEpidemiologyandBiostatisticsPlatform,OntarioInstituteforCancerResearch,Toronto,Ontario,Canada.
37ProssermanCentreforHealthResearch,SamuelLunenfeldResearchInstitute,Toronto,Ontario,Canada.
38CentreforGeneticEpidemiologyandBiostatistics,TheUniversityofWesternAustralia,Crawley,WesternAustralia,Australia.
39DepartmentofInternalMedicine,VUUniversityMedicalCentre,Amsterdam,TheNetherlands.
40UnitofPublicHealthGenomics,DepartmentofChronicDiseasePrevention,NationalInstituteforHealthandWelfare,Helsinki,Finland.
41CardiovascularMedicine,UniversityofOxford,WellcomeTrustCentreforHumanGenetics,Oxford,UK.
42DepartmentofBiologicalPsychology,VUUniversityAmsterdam,Amsterdam,TheNetherlands.
43MRCHumanGeneticsUnit,MRCInstituteforGeneticsandMolecularMedicine,WesternGeneralHospital,Edinburgh,UK.
44InstituteofMedicalBiometryandEpidemiology,UniversityofMarburg,Marburg,Germany.
45DepartmentofImmunology,GeneticsandPathology,UppsalaUniversity,Uppsala,Sweden.
46UppsalaClinicalResearchCenter,UppsalaUniversityHospital,Uppsala,Sweden.
47GenomeCentre,BartsandTheLondonSchoolofMedicineandDentistry,QueenMaryUniversityofLondon,London,UK.
48ClinicalPharmacology,WilliamHarveyResearchInstitute,BartsandTheLondonSchoolofMedicineandDentistry,QueenMaryUniversityofLondon,London,UK.
49LudwigshafenRiskandCardiovascularHealth(LURIC)Study,Freiburg,Germany.
50MannheimInstituteofPublicHealth,SocialandPreventiveMedicine,MedicalFacultyofMannheim,UniversityofHeidelberg,Mannheim,Germany.
51InstitutfürMedizinischeBiometrieundStatistik,UniversittzuLübeck,UniversittsklinikumSchleswig-Holstein,CampusLübeck,Lübeck,Germany.
52DepartmentofMedicalGenetics,UniversityofLausanne,Lausanne,Switzerland.
53SwissInstituteofBioinformatics,Lausanne,Switzerland.
54DepartmentofMedicalGenetics,DivisionofGeneticEpidemiology,MolecularandClinicalPharmacology,InnsbruckMedicalUniversity,Innsbruck,Austria.
55UniversityLilleNorddeFrance,Lille,France.
56CentreNationaldelaRechercheScientifique(CNRS)UnitéMixtedeRecherche(UMR)8199,InstitutdeBiologiedeLille(IBL)–InstitutPasteurdeLille,Lille,France.
57CardiovascularHealthResearchUnit,UniversityofWashington,Seattle,Washington,USA.
58DepartmentofTwinResearchandGeneticEpidemiology,King'sCollegeLondon,London,UK.
59SchoolofSocialandCommunityMedicine,UniversityofBristol,Bristol,UK.
60DepartmentofInternalMedicine,ErasmusMedicalCenter,Rotterdam,TheNetherlands.
61NetherlandsGenomicsInitiative(NGI)-sponsoredNetherlandsConsortiumforHealthyAging(NCHA),TheNetherlands.
62DepartmentofMedicineI,UniversityHospitalGrosshadern,Ludwig-Maximilians-Universitt,Munich,Germany.
63ChairofEpidemiology,InstituteofMedicalInformatics,BiometryandEpidemiology,Ludwig-Maximilians-Universitt,Munich,Germany.
64ChairofGeneticEpidemiology,InstituteofMedicalInformatics,BiometryandEpidemiology,Ludwig-Maximilians-Universitt,Munich,Germany.
65DepartmentofBiostatistics,BostonUniversitySchoolofPublicHealth,Boston,Massachusetts,USA.
66DepartmentofEpidemiology,UniversityofGroningen,UniversityMedicalCenterGroningen,Groningen,TheNetherlands.
67MRCCentreforCausalAnalysesinTranslationalEpidemiology,SchoolofSocialandCommunityMedicine,UniversityofBristol,Bristol,UK.
68InstituteforMedicalInformatics,BiometryandEpidemiology(IMIBE),UniversityHospitalofEssen,UniversityofDuisburg-Essen,Essen,Germany.
69InstituteforMolecularMedicineFinland(FIMM),UniversityofHelsinki,Helsinki,Finland.
70MedizinischeKlinikII,UniversittzuLübeck,Lübeck,Germany.
71DivisionofPreventiveMedicine,BrighamandWomen'sHospital,Boston,Massachusetts,USA.
72GeneticandMolecularEpidemiologyUnit,DepartmentofClinicalSciences,SkneUniversityHospitalMalm,LundUniversity,Malm,Sweden.
73DepartmentofPublicHealth&ClinicalMedicine,UmeUniversity,Ume,Sweden.
74DepartmentofOdontology,UmeUniversity,Ume,Sweden.
75IcelandicHeartAssociation,Kopavogur,Iceland.
76FacultyofMedicine,UniversityofIceland,Reykjavik,Iceland.
77AtherosclerosisResearchUnit,DepartmentofMedicine,Solna,KarolinskaInstitutet,KarolinskaUniversityHospital,Stockholm,Sweden.
78InterfacultyInstituteforGeneticsandFunctionalGenomics,Ernst-Moritz-Arndt-UniversityGreifswald,Greifswald,Germany.
79DepartmentofVascularMedicine,AcademicMedicalCenter,Amsterdam,TheNetherlands.
80HeartFailureResearchCentre,DepartmentofClinicalandExperimentalCardiology,AcademicMedicalCenter,Amsterdam,TheNetherlands.
81DepartmentofOncology,UniversityofCambridge,Cambridge,UK.
82DepartmentofEndocrinology,UniversityMedicalCenterGroningen,UniversityofGroningen,Groningen,TheNetherlands.
83LifeLinesCohortStudy,UniversityMedicalCenterGroningen,UniversityofGroningen,Groningen,TheNetherlands.
84DepartmentofInternalMedicine,InstituteofMedicine,SahlgrenskaAcademy,UniversityofGothenburg,Gothenburg,Sweden.
85HudsonAlphaInstituteforBiotechnology,Huntsville,Alabama,USA.
86DepartmentofCardiology,DivisionofHeart&Lungs,UniversityMedicalCenterUtrecht,Utrecht,TheNetherlands.
87InstituteofBiomedicine/Physiology,UniversityofEasternFinland,KuopioCampus,Kuopio,Finland.
88NationalInstituteforHealthResearch(NIHR)CambridgeBiomedicalResearchCentre,Cambridge,UK.
89DivisionofEpidemiology,MultidisciplinaryCardiovascularResearchCentre(MCRC),LeedsInstituteofGenetics,HealthandTherapeutics(LIGHT),UniversityofLeeds,Leeds,UK.
90DepartmentofMolecularGenetics,PathWestLaboratoryofWesternAustralia,QueenElizabethIIMedicalCentre,Nedlands,WesternAustralia,Australia.
91DepartmentofSurgeryandPathology,UniversityofWesternAustralia,Nedlands,WesternAustralia,Australia.
92GenomeTechnologyBranch,NationalHumanGenomeResearchInstitute,USNationalInstitutesofHealth,Bethesda,Maryland,USA.
93DipartimentodiMedicinaSperimentale.
UniversitàdegliStudiMilano-Bicocca,Monza,Italy.
94CentreforPopulationHealthSciences,UniversityofEdinburgh,Edinburgh,UK.
95HarvardMedicalSchool,Boston,Massachusetts,USA.
96BritishHeartFoundationGlasgowCardiovascularResearchCentre,UniversityofGlasgow,Glasgow,UK.
97UniversityofDundee,NinewellsHospital&MedicalSchool,Dundee,UK.
98NationalHeartandLungInstitute,ImperialCollegeLondon,London,UK.
99DivisionofCardiovascularEpidemiology,InstituteofEnvironmentalMedicine,KarolinskaInstitutet,Stockholm,Sweden.
100DepartmentofNatureGeneticsVOLUME45|NUMBER5|MAY2013511ArticlesEpidemiology,BiostatisticsandHTA,RadboudUniversityNijmegenMedicalCentre,Nijmegen,TheNetherlands.
101IstitutodiRicercaGeneticaeBiomedicadelConsiglioNazionaledelleRicerche(CNR),Monserrato,Italy.
102DepartmentofDietetics-Nutrition,HarokopioUniversity,Athens,Greece.
103EpidemiologyandPreventiveMedicineResearchCenter,DepartmentofClinicalandExperimentalMedicine,UniversityofInsubria,Varese,Italy.
104DepartmentofCardiology,ToulouseUniversitySchoolofMedicine,RangueilHospital,Toulouse,France.
105DepartmentofGenetics,UniversityMedicalCenterGroningen,UniversityofGroningen,Groningen,TheNetherlands.
106DepartmentofHealthSciences,UniversityofMilan,OspedaleSanPaolo,Milan,Italy.
107UniversityofChicago,Chicago,Illinois,USA.
108NorthshoreUniversityHealthSystem,Evanston,Illinois,USA.
109ResearchUnitforMolecularEpidemiology,HelmholtzZentrumMünchen–GermanResearchCenterforEnvironmentalHealth,Neuherberg,Germany.
110DivisionofCardiovascularandNeuronalRemodelling,MultidisciplinaryCardiovascularResearchCentre,LIGHT,UniversityofLeeds,Leeds,UK.
111DepartmentofClinicalSciences/ObstetricsandGynecology,UniversityofOulu,Oulu,Finland.
112DepartmentofNeurology,BostonUniversitySchoolofMedicine,Boston,Massachusetts,USA.
113DepartmentofPsychiatry,WashingtonUniversitySchoolofMedicine,St.
Louis,Missouri,USA.
114DepartmentofChildandAdolescentPsychiatry,UniversityofDuisburg-Essen,Essen,Germany.
115CentreForPaediatricEpidemiologyandBiostatistics,MRCCentreofEpidemiologyforChildHealth,UniversityCollegeLondonInstituteofChildHealth,London,UK.
116DivisionofResearch,KaiserPermanenteNorthernCalifornia,Oakland,California,USA.
117CoreGenotypingFacility,SAIC-Frederick,Inc.
,NationalCancerInstitute(NCI)-Frederick,Frederick,Maryland,USA.
118DepartmentofPhysiology,InstituteofNeuroscienceandPhysiology,SahlgrenskaAcademy,UniversityofGothenburg,Gothenburg,Sweden.
119PopulationStudiesUnit,DepartmentofChronicDiseasePrevention,NationalInstituteforHealthandWelfare,Turku,Finland.
120DepartmentofClinicalPhysiology,UniversityofTampereandTampereUniversityHospital,Tampere,Finland.
121CardiovascularResearchCenter,MassachusettsGeneralHospital,Boston,Massachusetts,USA.
122CardiologyDivision,MassachusettsGeneralHospital,Boston,Massachusetts,USA.
123CenterforHumanGeneticResearch,MassachusettsGeneralHospital,Boston,Massachusetts,USA.
124UKClinicalResearchCollaboration(UKCRC)CentreofExcellenceforPublicHealth(NorthernIreland),QueensUniversity,Belfast,UK.
125DepartmentofPublicHealthandPrimaryCare,InstituteofPublicHealth,UniversityofCambridge,Cambridge,UK.
126DepartmentofEpidemiologyandPublicHealth,UniversityCollegeLondon,London,UK.
127DepartmentofInternalMedicineII–Cardiology,UniversityofUlmMedicalCenter,Ulm,Germany.
128ChronicDiseaseEpidemiologyandPreventionUnit,DepartmentofChronicDiseasePrevention,NationalInstituteforHealthandWelfare,Helsinki,Finland.
129DepartmentofMedicine,UniversityofEasternFinland,KuopioCampusandKuopioUniversityHospital,Kuopio,Finland.
130FinnishInstituteofOccupationalHealth,Oulu,Finland.
131KuopioResearchInstituteofExerciseMedicine,Kuopio,Finland.
132LaboratoryofEpidemiology,Demography,Biometry,NationalInstituteonAging,USNationalInstitutesofHealth,Bethesda,Maryland,USA.
133DepartmentofMedicalSciences,UppsalaUniversity,AkademiskaSjukhuset,Uppsala,Sweden.
134DiabetesPreventionUnit,NationalInstituteforHealthandWelfare,Helsinki,Finland.
135HumanGenetics,GenomeInstituteofSingapore,Singapore.
136DepartmentofInternalMedicine,IstitutoAuxologicoItaliano,Verbania,Italy.
137TransplantationLaboratory,HaartmanInstitute,UniversityofHelsinki,Helsinki,Finland.
138ChairofNephrologySanRaffaeleScientificInstitute,UniversitàVita-SaluteSanRaffaele,UnitàOspedalieraNephrologyandDialysis,Milan,Italy.
139SynlabAcademy,Mannheim,Germany.
140DepartmentofCardiology,UniversityMedicalCenterGroningen,UniversityofGroningen,Groningen,TheNetherlands.
141DepartmentofBiostatistics,UniversityofWashington,Seattle,Washington,USA.
142GeneticsDivision,GlaxoSmithKline,KingofPrussia,Pennsylvania,USA.
143InstituteofHumanGenetics,UniversityofBonn,Bonn,Germany.
144DepartmentofGenomics,Life&BrainCenter,UniversityofBonn,Bonn,Germany.
145SchoolofPopulationHealth,TheUniversityofWesternAustralia,Nedlands,WesternAustralia,Australia.
146DepartmentofRespiratoryMedicine,SirCharlesGairdnerHospital,Nedlands,WesternAustralia,Australia.
147BusseltonPopulationMedicalResearchFoundation,SirCharlesGairdnerHospital,Nedlands,WesternAustralia,Australia.
148DepartmentofInternalMedicine,UniversityMedicalCenterGroningen,UniversityofGroningen,Groningen,TheNetherlands.
149MRCHarwell,Harwell,UK.
150DepartmentofStatistics,UniversityofOxford,Oxford,UK.
151SectionofEpidemiology,DepartmentofPublicHealth,AarhusUniversity,Aarhus,Denmark.
152MRCUnitforLifelongHealth&Ageing,London,UK.
153DepartmentofClinicalGenetics,ErasmusMedicalCenter,Rotterdam,TheNetherlands.
154CenterofMedicalSystemsBiology,LeidenUniversityMedicalCenter,Leiden,TheNetherlands.
155MedicalResearchInstitute,UniversityofDundee,NinewellsHospitalandMedicalSchool,Dundee,UK.
156Illumina,Inc.
,Cambridge,UK.
157InstituteofEpidemiologyII,HelmholtzZentrumMünchen–GermanResearchCenterforEnvironmentalHealth,Neuherberg,Germany.
158MunichHeartAlliance,Munich,Germany.
159FacultyofMedicine,UniversityofSplit,Split,Croatia.
160NationalInstituteforHealthandWelfare,Oulu,Finland.
161CenterforBiomedicine,EuropeanAcademyBozen/Bolzano(EURAC),Bolzano/Bozen,Italy(affiliatedinstituteoftheUniversityofLübeck).
162DepartmentofNeurology,GeneralCentralHospital,Bolzano,Italy.
163DepartmentofNeurology,UniversityofLübeck,Lübeck,Germany.
164OxfordCentreforDiabetes,EndocrinologyandMetabolism,UniversityofOxford,Oxford,UK.
165DepartmentofHaematology,UniversityofCambridge,Cambridge,UK.
166NationalHealthService(NHS)BloodandTransplant,CambridgeCentre,Cambridge,UK.
167ResearchCentreofAppliedandPreventiveCardiovascularMedicine,UniversityofTurku,Turku,Finland.
168TheDepartmentofClinicalPhysiologyandNuclearMedicine,TurkuUniversityHospital,Turku,Finland.
169MRCBiostatisticsUnit,InstituteofPublicHealth,Cambridge,UK.
170FinnishDiabetesAssociation,Tampere,Finland.
171PirkanmaaHospitalDistrict,Tampere,Finland.
172SouthKareliaCentralHospital,Lappeenranta,Finland.
173InstituteforCommunityMedicine,UniversityMedicineGreifswald,Greifswald,Germany.
174InstituteforClinicalMolecularBiology,Christian-AlbrechtsUniversity,Kiel,Germany.
175DeutschesHerzzentrumMünchenandDZHK(GermanCentreforCardiovascularResearch),partnersiteMunichHeartAlliance,Munich,Germany.
176AziendaOspedalieradiDesioeVimercate,Milan,Italy.
177DivisionofCardiology,CardiovascularLaboratory,HelsinkiUniversityCentralHospital,Helsinki,Finland.
178UniversityofEasternFinlandandKuopioUniversityHospital,Kuopio,Finland.
179KlinikundPoliklinikfürInnereMedizinII,UniversittklinikumRegensburg,Regensburg,Germany.
180DepartmentofMedicine,UniversityofLeipzig,Leipzig,Germany.
181IntegrierteForschungs-undBehandlungszentren(IFB)AdiposityDiseases,UniversityofLeipzig,Leipzig,Germany.
182InstitutNationaldelaSantéetdelaRechercheMédicale(INSERM)UnitéMixtedeRechercheScientifique(UMRS)937,InstituteofCardiometabolismAndNutrition(ICAN),PierreetMarieCurieMedicalSchool,Paris,France.
183DipartimentodiScienzeFarmacologicheeBiomolecolari,UniversitàdiMilano,CentroCardiologicoMonzino,IstitutodiRicoveroeCuraaCarattereScientifico(IRCCS),Milan,Italy.
184DepartmentofHumanGenetics,RadboudUniversityNijmegenMedicalCentre,Nijmegen,TheNetherlands.
185DepartmentofMedicine,UniversityofTurkuandTurkuUniversityHospital,Turku,Finland.
186DepartmentofInternalMedicine,CentreHospitalierUniversitaireVaudois(CHUV)UniversityHospital,Lausanne,Switzerland.
187CardiologyGroup,Frankfurt-Sachsenhausen,Germany.
188InstitutPasteurdeLille,INSERMU744,UniversitéLilleNorddeFrance,Lille,France.
189DepartmentofMedicine,DivisionofEndocrinologyandDiabetes,UniversityHospital,Ulm,Germany.
190HumanGeneticsCenter,UniversityofTexasHealthScienceCenter,Houston,Texas,USA.
191InstituteofMolecularMedicine,UniversityofTexasHealthScienceCenter,Houston,Texas,USA.
192FondazioneFilarete,Milan,Italy.
193DepartmentofGeneralPracticeandPrimaryHealthCare,UniversityofHelsinki,Helsinki,Finland.
194NationalInstituteforHealthandWelfare,Helsinki,Finland.
195UnitofGeneralPractice,HelsinkiUniversityCentralHospital,Helsinki,Finland.
196DepartmentofGenomicsofCommonDisease,SchoolofPublicHealth,ImperialCollegeLondon,London,UK.
197HelseunderskelseniNord-Trndelag(HUNT)ResearchCentre,DepartmentofPublicHealthandGeneralPractice,NorwegianUniversityofScienceandTechnology,Levanger,Norway.
198HannoverUnifiedBiobank,HannoverMedicalSchool,Hannover,Germany.
199DepartmentofEpidemiologyandBiostatistics,SchoolofPublicHealth,FacultyofMedicine,ImperialCollegeLondon,London,UK.
200InstituteofHealthSciences,UniversityofOulu,Oulu,Finland.
201BiocenterOulu,UniversityofOulu,Oulu,Finland.
202UnitofGeneralPractice,OuluUniversityHospital,Oulu,Finland.
203DepartmentofUrology,RadboudUniversityNijmegenMedicalCentre,Nijmegen,TheNetherlands.
204ComprehensiveCancerCenterEast,Nijmegen,TheNetherlands.
205DepartmentofClinicalChemistry,FimlabLaboratories,UniversityofTampereandTampereUniversityHospital,Tampere,Finland.
206StanfordUniversitySchoolofMedicine,Stanford,California,USA.
207DepartmentofClinicalMedicine,FacultyofHealthSciences,UniversityofTroms,Troms,Norway.
208DepartmentofCommunityMedicine,FacultyofHealthSciences,UniversityofTroms,Troms,Norway.
209InterdisciplinaryCenterPsychopathologyandEmotionRegulation,UniversityofGroningen,UniversityMedicalCenterGroningen,Groningen,TheNetherlands.
210DepartmentofPsychiatry,UniversityMedicalCentreGroningen,Groningen,TheNetherlands.
211DepartmentofEpidemiology,UniversityofWashington,Seattle,Washington,USA.
212DepartmentofMedicine,UniversityofWashington,Seattle,Washington,USA.
213DepartmentofHealthServices,UniversityofWashington,Seattle,Washington,USA.
214GroupHealthResearchInstitute,GroupHealth,Seattle,Washington,USA.
215DepartmentofClinicalPhysiologyandNuclearMedicine,KuopioUniversityHospital,Kuopio,Finland.
216DepartmentofCardiovascularSciences,UniversityofLeicester,GlenfieldHospital,Leicester,UK.
217LeicesterNIHRBiomedicalResearchUnitinCardiovascularDisease,GlenfieldHospital,Leicester,UK.
218InstituteofPreventiveMedicine,BispebjergUniversityHospital,Copenhagen,Denmark.
219NovoNordiskFoundationCenterforBasicMetabolicResearch,UniversityofCopenhagen,Copenhagen,Denmark.
220CentreforVascularPrevention,Danube-UniversityKrems,Krems,Austria.
221KingAbdulazizUniversity,Jeddah,SaudiArabia.
222InstituteofPublicHealthandClinicalNutrition,UniversityofEasternFinland,Kuopio,Finland.
223ResearchUnit,KuopioUniversityHospital,Kuopio,Finland.
224InstituteofClinicalChemistryandLaboratoryMedicine,UniversityMedicineGreifswald,Greifswald,Germany.
225InstituteofEpidemiologyI,HelmholtzZentrumMünchen–GermanResearchCenterforEnvironmentalHealth,Neuherberg,Germany.
226KlinikumGrosshadern,Munich,Germany.
227DepartmentofMedicine,StanfordUniversitySchoolofMedicine,Stanford,California,USA.
228UniversityofCambridgeMetabolicResearchLaboratories,InstituteofMetabolicScienceAddenbrooke'sHospital,Cambridge,UK.
229DivisionofIntramuralResearch,NationalHeart,Lung,andBloodInstitute,512VOLUME45|NUMBER5|MAY2013NatureGeneticsFraminghamHeartStudy,Framingham,Massachusetts,USA.
230LundUniversityDiabetesCentre,DepartmentofClinicalSciences,LundUniversity,Malm,Sweden.
231MedicalGeneticsInstitute,Cedars-SinaiMedicalCenter,LosAngeles,California,USA.
232DepartmentofEpidemiologyandPopulationHealth,AlbertEinsteinCollegeofMedicine,Bronx,NewYork,USA.
233OxfordNIHRBiomedicalResearchCentre,ChurchillHospital,Oxford,UK.
234DepartmentofMedicine,UniversityofMarylandSchoolofMedicine,Baltimore,Maryland,USA.
235DepartmentofGeneticsandGenomicSciences,MountSinaiSchoolofMedicine,NewYork,NewYork,USA.
236InstituteofGenomicsandMultiscaleBiology,MountSinaiSchoolofMedicine,NewYork,NewYork,USA.
237LaboratoryofGenetics,NationalInstituteonAging,Baltimore,Maryland,USA.
238DivisionofPopulationHealthSciencesandEducation,StGeorge's,UniversityofLondon,London,UK.
239DepartmentofClinicalEpidemiologyandBiostatistics,McMasterUniversity,Hamilton,Ontario,Canada.
240DepartmentofInternalMedicine,DivisionofGastroenterology,UniversityofMichigan,AnnArbor,Michigan,USA.
241TheCharlesBronfmanInstituteofPersonalizedMedicine,IcahnSchoolofMedicineatMountSinai,NewYork,NewYork,USA.
242TheMindichChildHealthandDevelopmentInstitute,IcahnSchoolofMedicineatMountSinai,NewYork,NewYork,USA.
243DepartmentofPreventiveMedicine,IcahnSchoolofMedicineatMountSinai,NewYork,NewYork,USA.
244Theseauthorscontributedequallytothiswork.
245Theseauthorsjointlydirectedthiswork.
CorrespondenceshouldbeaddressedtoE.
I.
(erik.
ingelsson@medsci.
uu.
se).
ArticlesNatureGeneticsdoi:10.
1038/ng.
2606ONLINEMETHODSStudydesign.
Weconductedatwo-stagestudyforthedistributiontailsofthreeanthropometrictraits(BMI,WHRadjustedforBMIandheight)andfourclinicalclassesofobesity(overweightclassandobesityclassesI,IIandIII),followedbyacombinedanalysisofthetwostages.
Stage1consistedofameta-analysisofGWASusingdatafromastudybase(orsamplingframe)ofupto168,267adultindividualsofEuropeanancestryfrom51studiesparticipatingintheGIANTConsortium(SupplementaryTables1–5).
Instage2,273SNPswithPvalue<5*106werefollowedupinupto109,703additionalindividualsofEuropeandescent,whoincluded67,243individualsfrom24studieswithdatafromtheMetabochip(acustom-designedarrayof~200,000SNPswithpreviousevidenceofsuggestiveassociationwithmeta-bolictraits)and42,460individualsfrom12studieswithinsilicoreplicationGWASdata(SupplementaryTables1–5).
Thisgaveusastudybaseofupto276,007individualsofEuropeandescentforthejointmeta-analysisofstage1andstage2.
Fulldetailsofthediscoveryandreplicationstages,analysisofdataandmeta-analysesaregivenintheSupplementaryNote.
Allstudieswereapprovedbylocalresearchethicscommittees,andallparticipantsgaveinformedconsent.
Phenotypedefinitions.
Thedistributiontailsofthethreeanthropometrictraits(BMI,WHRadjustedforBMIandheight)weredefinedastheupper5thpercentile(cases)andlower5thpercentile(controls)ofthedistributionstratifiedbysexanddiseasestatusaftercontrollingforthefollowingcovariates:age,age2andprincipalcomponentsforBMI;ageandprincipalcomponentsforheight;andage,age2,BMIandprincipalcomponentsforWHR.
Fortheclini-calobesityclasses,casesweredefinedbyBMI≥25kg/m2fortheoverweightclass,BMI≥30kg/m2forobesityclassI,BMI≥35kg/m2forobesityclassIIandBMI≥40kg/m2forobesityclassIII.
ControlsweresubjectswithBMI<25kg/m2.
Aminimumof30casesand30controlsforeachstudy-specificstratumwasrequired.
Associationanalysesandmeta-analyses.
Eachstudyconductedsingle-markerassociationanalysesassuminganadditivegeneticmodeltakinggenotypeimpu-tationuncertaintyintoaccount.
Analyseswerestratifiedbysex(exceptforstud-ieswithrelatedindividuals,whereanalysesaccountedforfamilystructure)anddiseasestatusforstudiesthatascertainedparticipantsonthebasisofarelevantdisease(forexample,diabetes).
Beforecarryingoutmeta-analysisofthedata,theresultsfromeachstudywereextensivelyreviewedusingstandardizedqualitycontrolprocedurestoidentifypotentialproblems,suchasstrandissues,discrep-anciesbetweenthereportedstandarderrorsandPvaluesandallelefrequencydifferences.
SNPswithpoorimputationqualityscoresorestimatedminorallelecount≤20(2*N*minorallelefrequency)ineachstratum(men/womenorpooledforfamily-basedstudies)ofeachstudywereremovedfromanalysis.
Fordiscoverystage1,eachstratum-andstudy-specificGWASwascorrectedforgenomiccontrol.
Meta-analyseswereperformedforeachphenotypeinMETAL48usingthefixed-effectsinversevariancemethodbasedonβestimatesandstandarderrorsfromeachstudy.
Theresultsofthediscoverymeta-analysiswerecorrectedbyanadditionalgenomiccontrolstep.
Similarmethodswereemployedforthereplicationandjointdiscoveryandreplicationanalyses.
Associationtestinginstudiesofextremelyobeseindividuals.
WetestedtheassociationofallSNPsreachingP<5*108inthejointanalysisofstage1andstage2resultsfortheBMI-relatedtraitsinfivestudiesofextremelyobeseindividuals(SupplementaryTables2–5).
Forthefourcase-controlstudies(FrenchExtremeObesityStudy,EssenCase-ControlGWAS,GEOandGOYA),afixed-effectsinversevariancemethodwasusedtoperformmeta-analysisoftheresults.
Withthefourcase-controlstudies,meta-analysiswasperformedincludingafifthstudy(EssenObesityTrioGWAS)thathasanuclearstructure,usingaweightedz-scoremethodthattakesintoaccountthedirectionbutnotthemagnitudeoftheassociation.
Systematiccomparisonofthegeneticstructurebetweendistributiontailsandtheoveralldistribution.
Fortheseanalyses,weincludedallGWASthatprovidedgenome-wideresultsforboththefulldistributionandthedistribu-tiontailsofBMI,heightandWHR.
First,weusedtheresultsforthefulldis-tributiontocalculate,foreachgenotype,theexpectednumberofindividualsintheupperandlower5%tails.
Weusedthesevaluestoperformlogisticregression,comparingtheupperandlowertails,andobtainedtheexpectedβvaluesandexpectedstandarderror.
Second,wetestedthedifferencesbetweentheexpectedβvaluesandtheobservedβvaluesobtainedfromthemeta-analysesofthetailsofthedistribu-tions.
Thestandarderroroftheeachdifferencewasestimatedassqrt((expectedstandarderror)2+(observedstandarderror)22*0.
65*(expectedstand-arderror*observedstandarderror)),where0.
65isthecorrelationbetweentheexpectedβvaluesandtheobservedβvaluesobtainedfromTWINGENEbybootstrapping.
Finally,meta-analysiswasperformedonthedifferencesbetweenexpectedβvaluesandobservedβvaluesusingtheinversevariancemethodinMETAL.
PolygeniccomparisonofgeneticdeterminantsoftheBMIdistributiontailsandoveralldistribution.
Withineachtrait(BMI,heightandWHR),weaimedtocomparevarianceexplainedinthedistributiontailsofthetraitbytwogenetic(polygenic)scoresobtainedfrom(i)themeta-analysesofthedistributiontailsofthetraitand(ii)themeta-analysesofthefulldistribution.
Tomakethescorescomparable,welimitedpolygenicscoreconstructiontothestudiesthatprovidedgenome-widemeta-analysisresultsforbothdistribu-tiontailsandtheoveralldistribution.
AfterLDfiltering(usingr2≥0.
05and1Mbofdistance)andexcludingSNPspresentin<50%ofsamples,wecreatedpolygenicscoresastheweightedsumofriskallelesusingthemethodpro-posedbytheInternationalSchizophreniaConsortium45.
FortheBMIanaly-sis,theassociationbetweenthepolygenicscoresandthedistributiontailsofBMIwasinvestigatedinfoursamplesofextremelyobeseindividualsandtwoindependentcohortstudiesusingthesamedefinitionofdistributiontails(SupplementaryTable13).
OnlythetwoindependentcohortswereusedfortheheightandWHRanalyses.
Toestimatethephenotypicvarianceexplained,wefitlogisticorlinearregressionmodelsincludingage,sex,study-specificcovariatesandthepolygenicscoreaspredictorsandthedistributiontailsofthetraitortheoveralltraitasoutcomesinseparatemodels.
ThephenotypicvarianceexplainedbythepolygenicscoreswasdefinedasthedifferenceinR2(linearregression)orNagelkerkeR2(logisticregression)betweenthesemodelsandabasicmodelincludingonlyage,sexandstudy-specificcovari-atesaspredictors.
Secondarysignalsanalysis.
Toidentifypotentialsecondarysignals,weusedtheapproximateconditionalandjointanalysispreviouslyproposed46,whichusessummary-levelstatisticsandtheLDstructurefromareferencesampletoapproximateconditionalPvalues.
Themeta-analysisresultsforeachtraitwereanalyzedseparatelywithLDcorrectionbetweenSNPsestimatedfrom6,654unrelatedindividualsfromtheARICcohort.
Haplotype-basedanalyses.
Usingdatafromtenofthelargeststudies,wetestedtheassociationbetweenthedistributiontailsofheight,BMIandWHRandhaplotypesacrosseachestablishedandnewlocusseparatelyformalesandfemaleswithineachstudyusingGENEBPM49.
HaplotypeswereestimatedfromGWASSNPdatabymeansofanexpectation-maximizationalgorithmandthenclusteredaccordingtotheirallelicsimilarity.
Withinalogisticregres-sionmodelingframework,haplotypeswithinthesameclusterwereassignedthesamealleliceffect,reducingtherequirednumberofparameters.
Markov-chainMonteCarlotechniqueswereemployedtosampleoverthespaceofhaplotypeclustersandregressionmodelparameters.
Evidenceinfavorofahaplotypeassociationwiththetraitwasassessedbysumminglog10(Bayesfactors)acrossstudies.
Expressionquantitativetraitlocus(eQTL)analyses.
WeexaminedthecisassociationsbetweenSNPsthatreachedgenome-widesignificance(P<5*108)andtheexpressionofnearbygenesinmultipletissuesfrom5studiesdescribedpreviously:(i)subcutaneousadiposetissue(n=603)andwholeblood(n=747)fromdeCODE50;(ii)lymphoblastoidcelllines(n=830)fromachildhoodasthmastudy51;(iii)liver(n=707),subcutaneousfat(n=870)andomentalfat(n=916)tissuefromabariatricsurgerystudy52;(iv)subcutaneousabdominal(n=52)andgluteal(n=62)adiposetissueandwholeblood(n=65)fromMolOBB53;and(v)corticalbraintissue(n=193)fromasurveystudy54.
SNPsweretestedforcisassociationswithtranscriptsNatureGeneticsdoi:10.
1038/ng.
2606within500kbor1Mb,assuminganadditiveeffectoftheBMI-relatedalleleorusinganANOVAtestwithstudy-specificP-valuethresholdsusedtoaccountformultipletesting.
Conditionalanalyseswereperformedforallexpres-siondata,exceptforcorticaltissue,byconditioningthetrait-associatedSNPonthemostsignificantcis-associatedSNPfortheparticulargenetranscriptandviceversa.
AlistofSNPsidentifiedinotherGWASnearnewlociisgiveninSupplementaryTable17.
48.
Willer,C.
J.
,Li,Y.
&Abecasis,G.
R.
METAL:fastandefficientmeta-analysisofgenomewideassociationscans.
Bioinformatics26,2190–2191(2010).
49.
Morris,A.
P.
DirectanalysisofunphasedSNPgenotypedatainpopulation-basedassociationstudiesviaBayesianpartitionmodellingofhaplotypes.
Genet.
Epidemiol.
29,91–107(2005).
50.
Emilsson,V.
etal.
Geneticsofgeneexpressionanditseffectondisease.
Nature452,423–428(2008).
51.
Dixon,A.
L.
etal.
Agenome-wideassociationstudyofglobalgeneexpression.
Nat.
Genet.
39,1202–1207(2007).
52.
Zhong,H.
,Yang,X.
,Kaplan,L.
M.
,Molony,C.
&Schadt,E.
E.
Integratingpathwayanalysisandgeneticsofgeneexpressionforgenome-wideassociationstudies.
Am.
J.
Hum.
Genet.
86,581–591(2010).
53.
Min,J.
L.
etal.
Coexpressionnetworkanalysisinabdominalandglutealadiposetissuerevealsregulatorygeneticlociformetabolicsyndromeandrelatedphenotypes.
PLoSGenet.
8,e1002505(2012).
54.
Myers,A.
J.
etal.
Asurveyofgenetichumancorticalgeneexpression.
Nat.
Genet.
39,1494–1499(2007).