人类群体遗传学—第二讲

人类群体遗传学基本原理和分析方法中科院-马普学会计算生物学伙伴研究所中国科学院上海生命科学研究院研究生课程人类群体遗传学徐书华金力序号日期课程内容授课教师12月26日Hardy-Weinberg平衡检验原理及其应用徐书华23月5日遗传多态性统计量徐书华33月12日进化树的构建方法及应用徐书华43月19日Coalescence原理及应用李海鹏53月26日遗传漂变效应及有效群体大小的估计徐书华64月2日人群遗传结构分析（I）徐书华74月9日单倍型估计及连锁不平衡分析徐书华84月16日人群遗传结构分析（II）徐书华94月23日基因定位中的关联分析(I)何云刚104月30日基因定位中的关联分析(II)徐书华115月7日人类基因组中的连锁不平衡模式及标签位点的选择徐书华125月14日基因表达数据的分析方法严军135月21日人群历史的遗传学研究徐书华5月28日端午节146月4日法医学检测及分析方法李士林156月11日自然选择检验原理和方法徐书华166月18日全基因组基因型数据正选择检验方法徐书华176月25日课程考试教育基地2008－2009学年第二学期《人类群体遗传学分析方法》课程表上课时间：每周四上午10:00-11:50上课地点：中科大厦4楼403室第7教室第二讲遗传多态性统计量第二讲遗传多态性的概念遗传多态性的种类描述遗传多态性的统计量群体遗传多态性参数（θ）的估计利用群体遗传多态性数据进行统计检验TajimatestPolymorphismLight-morphJaguar(typical)Dark-morphormelanisticJaguar(about6%oftheSouthAmericanpopulation):1100(2008)PolymorphismGreek:poly=many,andmorph=formPolymorphismisoftendefinedasthepresenceofmorethanonegeneticallydistincttypeinasinglepopulation.Rarevariationsarenotclassifiedaspolymorphisms;andmutationsbythemselvesdonotconstitutepolymorphisms.SexualdimorphismWhyistheratio~50/50?DNApolymorphismRFLP(RestrictionFragmentLengthPolymorphism)AFLP(AmplifiedFragmentLengthPolymorphism)RAPD(RandomAmplificationofPolymorphicDNA)VNTR(VariableNumberTandemRepeat,orMinisatellite)STR(ShortTandemRepeat,orMicrosatellite)SNP(SingleNucleotidePolymorphism)SFP(SingleFeaturePolymorphism)CNV(CopyNumberVariation)IntuitivestatisticsNumberofallelesMorealleles,largerdiversity;Minorallelefrequency(MAF)isthefrequencyoftheless(orleast)frequentalleleinagivenlocusandagivenpopulation.HumanSNPdataASingleNucleotidePolymorphism(SNP)(snip)isasinglebasevariantinDNA.Mutation:minorallelefrequency(MAF)≤1%SNP:MAF1%SNPsarethemostsimpleformandmostcommonsourceofgeneticpolymorphisminthehumangenome(90%ofallhumanDNApolymorphisms).HeterozygosityThefractionofindividualsinapopulationthatareheterozygousforaparticularlocus.Itcanalsorefertothefractionoflociwithinanindividualthatareheterozygous.wherenisthenumberofindividualsinthepopulation,andai1,ai2aretheallelesofindividualiatthetargetlocus.Observedwheremisthenumberofallelesatthetargetlocus,andfiistheallelefrequencyoftheithalleleatthetargetlocus.ExpectedHeterozygosityrelatedissuesHeterozygosityandHWDComparisonofHoandHeGenediversityPopulationMutationRate(q)Undermutation-driftequilibrium:q=4Nemforautosomeq=NemforYandmtDNAq=3NemforXchromosomeqautosomeqXqYEstimatorsofθNumberofsegregatingsites(θK);Averagepairwisedifferences(θ∏);Numberofalleles(θE);MeannumberofmutationssincetheMRCA(θΩ);Singleton.►Undertheinfinitesitemodel,Kisequaltothenumberofmutationssincethemostrecentcommonancestorofthesequencesinthesample.►Therefore,Khasaclearbiologicalmeaning.►However,Kdependsonthesamplesize.Numberofsegregatingsites(K)Kq►NormalizedKKnKaq11121nanKEqq22nKnnbVaraaqqq211141nbn►UndertheneutralWright-Fishermodelwithconstanteffectivepopulationsize,ThepropertiesofθKθKisindependentofsamplesize.However,theusefulnessofθKisnotclearunderotherpopulationgeneticmodels,suchasthosewithnaturalselection.θKissensitivetothenumberofrarealleles,ormutantsoflowfrequency.HowmanycommonSNPsinhumangenome?CommonSNPs:minorallelefrequency(MAF)0.05;Supposewehave50samplesofAfrican,European,Asianrespectively;Theta=1.2/kbforAfricanpopulation;Theta=0.8/kbforEuropeanandAsianpopulation;Autosomelength(L)=2.68billionbp;►Weexpect9.8millioncommonSNPsin50Africansamples;►Weexpect6.5millioncommonSNPsin50Europeansamples;►Weexpect6.5millioncommonSNPsin50Asiansamples;11111niiKniiq1045%61MAFKiESLiqwhereThetaK=1.2/kbThetaK=0.8/kbAveragepairwisedifferences(∏)Alsoknownassequencediversitymeannumberofnucleotidedifferencesbetweentwosequences.21ijijdnn22,2313191EnnnVarnnnqqqThepropertiesof∏∏asameasureofgeneticvariationhasclearbiologicalmeaningswhichdonotdependontheunderlyingevolutionaryprocess.IncomparisontoθK,itisinsensitivetotherarealleles,ormutantsoflowfrequency.∏isanusefulmeasureofpersistentgeneticvariation,andneutralgeneticvariationwhenpurifyingselectionisoperating.However,becauseitsvarianceisconsiderablylargerthanthatofθK,itisnotasgoodasθKforneutrallocus.Locus(length)p(x10-4)q(x10-4)m(x10-9)NeReferenceAPOE(5.5kb)5.36.87(S)23.57,300Fullertonetal.2000Chr.1(10kb)5.89.51(S)14.816,000Yuetal.2001Chr.22(10kb)8.813.2(S)2314,400Zhaoetal.2000Xchr.(10.2kb)3.66.8(S)18.412,300Kaessmannetal.1999Xchr.(4.2kb))-4.41(ML)19.27,700Harris&Hey1999Ychr.(64kb)0.742.01(S)24.88,100Thomsonetal.2000mtDNA(15.4kb)2828(p)3408,200Ingmanetal.2000Aluinsertions---17,500Sherryetal.1997NucleotideDiversityNumberofallelesEwens(1972)showsthatundertheinfiniteallelemodel111Eknqqqq►AnestimateofθcanbeobtainedbyresolvingtheaboveequationforθwithE(k)replacedbyk.TheestimateisknownasEwens’sestimatorθE.ThepropertiesofθEUndertheinfiniteallelemodel,θEisaboutthebestestimatoronecandevise.However,θEisslightlyupwardbiasedestimatorparticularlywhenθislarge.MeannumberofmutationssincetheMRCA(Ω)ThemeannumberΩofmutatio