资源描述:
《Family based association studies for NGS AJHG 2012》由会员上传分享,免费在线阅读,更多相关内容在学术论文-天天文库。
1、ARTICLEFamily-BasedAssociationStudiesforNext-GenerationSequencingYunZhu1andMomiaoXiong1,*Anindividualsdiseaseriskisdeterminedbythecompoundedactionofbothcommonvariants,inheritedfromremoteancestors,thatsegregatedwithinthepopulationandrarevariants,inheritedfromrecentancestors,thatsegregatedmainl
2、ywithinpedigrees.Next-generationsequencing(NGS)technologiesgeneratehigh-dimensionaldatathatallowanearlycompleteevaluationofgeneticvariation.Despitetheirpromise,NGStechnologiesalsosufferfromremarkablelimitations:higherrorrates,enrichmentofrarevariants,andalargeproportionofmissingvalues,aswella
3、sthefactthatmostcurrentanalyticalmethodsaredesignedforpopulation-basedassociationstudies.TomeettheanalyticalchallengesraisedbyNGS,weproposeageneralframeworkforsequence-basedassociationstudiesthatcanusevarioustypesoffamilyandunrelated-individualdatasampledfromanypopulationstructureandauniversa
4、lprocedurethatcantransformanypopulation-basedassociationteststatisticforuseinfamily-basedassociationtests.Wedevelopfamily-basedfunc-2tionalprincipal-componentanalysis(FPCA)withorwithoutsmoothing,ageneralizedT,combinedmultivariateandcollapsing(CMC)method,andsingle-markerassociationteststatisti
5、cs.Throughintensivesimulations,wedemonstratethatthefamily-basedsmoothedFPCA(SFPCA)hasthecorrecttypeIerrorratesandmuchmorepowertodetectassociationof(1)commonvariants,(2)rarevariants,(3)bothcommonandrarevariants,and(4)variantswithoppositedirectionsofeffectfromotherpopulation-basedorfamily-based
6、associationanalysismethods.Theproposedstatisticsareappliedtotwodatasetswithpedigreestructures.TheresultsshowthatthesmoothedFPCAhasamuchsmallerpvaluethanotherstatistics.Introductionrarevariantsarisefromrecentmutationsinpedi-8,10,11grees.GiventhatanindividualrarevariantwouldResequencingofexomes
7、andultimately,wholegenomeshavearelativelysmallimpactonthecommondiseasegenerateunprecedentedlymassive,high-dimensionalandthatrarevariantshaveverylowpopulationfrequen-genetic-variationdatathatallowanearlycompletecies,thepowerthecurrentanalytica