LETTERS
©2006 Nature Publishing Group doc.guandang.net edmixed-modelmethodforassociationmappingthataccountsformultiplelevelsofrelatednessJianmingYu1,9,GaelPressoir1,9,WilliamHBriggs2,IrieVrohBi1,MasanoriYamasaki3,JohnFDoebley2,MichaelDMcMullen3,4,BrandonSGaut5,DahliaMNielsen6,JamesBHolland4,7,StephenKresovich1,8&EdwardSBuckler1,4,8Aspopulationstructurecanresultinspuriousassociations,ithasconstrainedtheuseofassociationstudiesinhumanandplantgenetics.Associationmapping,however,holdsgreatpromiseiftruesignalsoffunctionalassociationcanbeseparatedfromthevastnumberoffalsesignalsgeneratedbypopulationstructure1,2.Wehavedevelopedauni edmixed-modelapproachtoaccountformultiplelevelsofrelatednesssimultaneouslyasdetectedbyrandomgeneticmarkers.Weappliedthisnewapproachtotwosamples:afamily-basedsampleof14humanfamilies,forquantitativegeneexpressiondissection,andasampleof277diversemaizeinbredlineswithcomplexfamilialrelationshipsandpopulationstructure,forquantitativetraitdissection.OurmethoddemonstratesimprovedcontrolofbothtypeIandtypeIIerrorratesoverothermethods.Asthisnewmethodcrossestheboundarybetweenfamily-basedandstructuredassociationsamples,
itprovidesapowerfulcomplementtocurrentlyavailable


methodsforassociationmapping.
Populationstructureisuniversalamongorganisms3,4.Itcanarise
naturallyintheformofherds,colonies,ethnicgroupsorothertypes
ofaggregations,owingtogeography,naturalselectionorarti cial
Population structureselection.Forassociationmapping,agivensamplemayfallintooneof vecategoriesde nedbypopulationstructureassociatedwithlocaladaptationordiversifyingselectionandfamilialrelatednessfromrecentcoancestry(Fig.1).Ideally,sampleswithminimalpopulationstructureorfamilialrelatedness(areaI)resultinthegreateststatisticalpower,providedthatthetraitofinterestiswelldistributed(Fig.1).Suchsamples,however,oftenproveverydif culttocollect,aresmallinsizeand/orhaveanarrowgeneticbasis.Family-basedsamples(areaII)havebeenexploitedtoavoidtheeffectofpopulationstructure5–8,butthesesamplesarealsolimitedbysamplesizeandallelicdiversityandcanbedif culttocollect,particularlyforlate-onsethumandiseases(Fig.1).Forquantitativetraits,theQuantitativeTransmissionDisequilibriumTest(QTDT)isonemethodwidelyusedforassocia-tionmappingwiththesefamily-basedsamples7.Samplesofincreasedsizeaswellasbroaderallelicdiversityinaspeciesoftencontainpopulationstructure(areaIV)orincludefamilialrelationshipswithinstructuredpopulation(areaIII;Fig.1).Forthesesamples,StructuredFigure1Differenttypesofsamplesusedforassociationmapping.Althoughallindividualsarerelatedviaalargegenealogicaltree,thepopulationstructureaxisdepictsrelationshipsamongmajorsubpopulationsassociatedwithlocaladaptationordiversifyingselection.Thefamilialrelatednessaxisdepictstherelationshipsamongindividualsfromrecentcoancestry.The
dottedlinesindicatethatcleardelineationsmaynotexistbetweenareas.
Methodslistedcanbeappliedtodifferentsampletypes,althoughtheyare
designedforspeci cpurposes(forexample,SAwasdesignedforsamples
withpopulationstructure,andTDTandQTDTweredesignedforfamily-based
samples).Thesemethodsdealwiththespeci csamplestructureeither
directly(asinSA,QTDT,TDTandthemixedmodel)orindirectlyby
adjustingtheteststatisticsfromregressionanalysis(asinGC).
1InstituteFamilial relatednessforGenomicDiversity,CornellUniversity,Ithaca,NewYork14853,USA.2DepartmentofGenetics,UniversityofWisconsin,Madison,Wisconsin53706,USA.ofPlantSciences,UniversityofMissouri,Columbia,Missouri65211,USA.4UnitedStatesDepartmentofAgriculture-AgriculturalResearchService
5(USDA-ARS).DepartmentofEcologyandEvolutionaryBiology,UniversityofCalifornia,Irvine,California92697,USA.6DepartmentofStatisticsand7DepartmentofCropScience,NorthCarolinaStateUniversity,Raleigh,NorthCarolina27695,USA.8DepartmentofPlantBreedingandGenetics,CornellUniversity,Ithaca,NewYork14853,USA.9Theseauthorscontributedequallytothiswork.CorrespondenceshouldbeaddressedtoE.S.B.(esb33@cornell.edu).3Division
Received7July;accepted13October;publishedonline25December2005;doi:10.1038/ng1702
NATUREGENETICS
A unified mixed-model me... 暂无评价 7页 2财富值 The Unified Test Method...Two classes are in USE association (B uses A) if instances of B can ...
cient mixed-model association (EMMA), which corrects for population structure and genetic relatedness in model organism association mapping. Our method takes ...
10 A?unified ifi d?mixed i d‐model d l?method th d?for f ? association?mapping pp g y???=???Xβ +???Sα +???Qv +???Zu + e Backgr...
A Grammar Analysis Model for the Unified Multimedia Query Language_专业资料 暂无评价
Towards a unified physic... 暂无评价 10页 免费 A unified mixed-model me...(Gerstner and Kistler, 2002), categorization and association, while avoiding ...
Association mapping for ... 暂无评价 14页 免费 Association mapping for .....‘Exploiting Genome-wide Association in Oilseed Brassicas: a model for ...
A unified mixed-model ... 暂无评价 7页 1下载券 Towards a generic mode... 暂无评价 16页 免费 A New Method for Chara... 暂无评价 4页 2.00 ...
ed mixed-model method for association mapping that account for multiple level of relatedness. Nat Genet 38:203–208 Zanetti S, Winzeler M, Keller M, ...
Customer Requirements Mapping Method Based on Association Rule Mining for Mass...A Method of Rotating M... 1人阅读 8页 2.00 Subsystem model-based ...
Association mapping of yield and its components in rice 2007_电子/电路_工程...ed mixed-model method for association mapping that accounts for multiple ...
热门文档