Normalization

《Normalization》由会员分享，可在线阅读，更多相关《Normalization（24页珍藏版）》请在装配图网上搜索。

1、DNA Microarray Bioinformatics-#27612NormalizationGetting the numbers comparableDNA Microarray Bioinformatics-#27612Sample PreparationHybridizationArray designProbe designQuestionExperimental DesignBuy Chip/ArrayStatistical AnalysisFit to Model(time series)Expression IndexCalculationAdvanced Data Ana

2、lysisClusteringPCAClassification Promoter AnalysisMeta analysisSurvival analysisRegulatory NetworkComparableGene Expression DataNormalizationImage analysisThe DNA Array Analysis PipelineDNA Microarray Bioinformatics-#27612Expression intensities are not just target concentrationsSample contaminationR

3、NA qualitySample preparationDye effect(cy3/cy5)Probe affinityHybridizationUnspecific signal(background)SaturationSpottingOther issues related to array manufacturingImage segmentationArray spatial effectsDNA Microarray Bioinformatics-#27612Gene-specific variationSpotting(size and shape)Cross-hybridiz

4、ationDyeBiological variationEffectNoiseGlobal variationRNA qualitySample preparationDyeHybridizationPhotodetection Systematic Two kinds of variation in the signalStochasticDNA Microarray Bioinformatics-#27612Gene-specific variation:Too random to be explicitly accounted for“noise”Global variation:Sim

5、ilar effect on many measurements Corrections can be estimated from dataNormalizationStatistical testing Sources of variationSystematic StochasticDNA Microarray Bioinformatics-#27612Calibration=Normalization=ScalingDNA Microarray Bioinformatics-#27612Nonlinear normalizationDNA Microarray Bioinformati

6、cs-#27612Lowess NormalizationOne of the most commonly utilized normalization techniques is the LOcally Weighted Scatterplot Smoothing(LOWESS)algorithm.MA*DNA Microarray Bioinformatics-#27612The Qspline methodFrom the empirical distribution,a number of quantiles are calculated for each of the channel

7、s to be normalized(one channel shown in red)and for the reference distribution(shown in black)A QQ-plot is made and a normalization curve is constructed by fitting a cubic spline functionAs reference one can use an artificial“median array”for a set of arrays or use a log-normal distribution,which is

8、 a good approximation.DNA Microarray Bioinformatics-#27612Once againqsplineWhen many microarrays are to be normalized to each other an average array can be used as targetAccumulating quantilesDNA Microarray Bioinformatics-#27612Invariant set normalization(Li and Wong)A invariant set of probes is use

9、d-Probes that does does not change intensity rank between arrays-A piecewise linear median line is calculated-This curve is used for normalizationDNA Microarray Bioinformatics-#27612Spatial biasestimateSpatial normalizationAfter intensitynormalizationAfter spatialnormalizationRaw dataAfter intensity

10、normalizationAfter intensitynormalizationAfter spatialnormalizationAfter spatialnormalizationDNA Microarray Bioinformatics-#27612Sample PreparationHybridizationArray designProbe designQuestionExperimental DesignBuy Chip/ArrayStatistical AnalysisFit to Model(time series)Expression IndexCalculationAdv

11、anced Data AnalysisClusteringPCAClassification Promoter AnalysisMeta analysisSurvival analysisRegulatory NetworkComparableGene Expression DataNormalizationImage analysisThe DNA Array Analysis PipelineDNA Microarray Bioinformatics-#27612Expression index valueSome microarrays have multiple probes addr

12、essing the expression of the same targetAffymetrix GeneChips have 11-20 probe pairs pr.Gene-Perfect Match(PM)-MisMatch(MM)PM:CGATCAATTGCACTATGTCATTTCT MM:CGATCAATTGCAGTATGTCATTTCTHowever for downstream analysis we often want to deal with only one value pr.gene.Therefore we want to collapse the inten

13、sities from many probes into one value:a gene expression index valueDNA Microarray Bioinformatics-#27612Expression index calculationSimplest method?MedianBut more sophisticated methods exists:dChip,RMA and MAS 5DNA Microarray Bioinformatics-#27612dChip(Li&Wong)Model:PMij=ij+eijOutlier removal:Identi

14、fy extreme residualsRemoveRe-fitIterateDistribution of errors eij assumed independent of signal strength(Li and Wong,2001)DNA Microarray Bioinformatics-#27612RMARobust Multi-array Average(RMA)expression measure(Irizarry et al.,Biostatistics,2003)For each probe set,re-write PMij=ij as:log(PMij)=log(i

15、)+log(j)Fit this additive model by iteratively re-weighted least-squares or median polishDNA Microarray Bioinformatics-#27612MAS.5MicroArray Suite version 5 usesSignal=TukeyBiweightlog(PMj-MM*j)MM*is an adjusted MM that is never bigger than PMTukey biweight is a robust average procedure with weights

16、 and outlier rejectionDNA Microarray Bioinformatics-#27612Std Dev of gene measures from 20 replicate arraysMethods compared on expression varianceStandard deviation of gene measures from 20 replicate arraysRMA:Blue and RedMAS5:GreendChip:BlackExpression levelFrom Terry speedDNA Microarray Bioinforma

17、tics-#27612RobustnessMAS5.0(Irizarry et al.,Biostatistics,2003)MAS 5.0Log fold change estimate from 1.25ug cRNALog fold change estimate from 20ug cRNADNA Microarray Bioinformatics-#27612RobustnessdChip(Irizarry et al.,Biostatistics,2003)dChipLog fold change estimate from 1.25ug cRNALog fold change e

18、stimate from 20ug cRNADNA Microarray Bioinformatics-#27612RobustnessRMA(Irizarry et al.,Biostatistics,2003)RMALog fold change estimate from 1.25ug cRNALog fold change estimate from 20ug cRNADNA Microarray Bioinformatics-#27612All of this is implemented inRIn the BioConductor packages affy(Gautier et

19、 al.,2003).DNA Microarray Bioinformatics-#27612ReferencesLi and Wong,(2001).Model-based analysis of oligonucleotide arrays:Model validation,design issues and standard error application.Genome Biology 2:111.Irizarry,Bolstad,Collin,Cope,Hobbs and Speed,(2003)Summaries of Affymetrix GeneChip probe level data.Nucleic Acids Research 31(4):e15.)Affymetrix.Affymetrix Microarray Suite User Guide.Affymetrix,Santa Clara,CA,version 5 edition,2001.Gautier,Cope,Bolstad,and Irizarry,(2003).affy-an r package for the analysis of affymetrix genechip data at the probe level.Bioinformatics