?Dear All, I carried out a time series analysis (4 time points with a common reference) ?analysis of my dataset using LIMMA method. I set the design and contrast?matrix as: ? design<-model.matrix(~-1+factor(c(1,1,1,2,2,2,3,3,3,3,4,4,4,5,5,5,5,5) )) > colnames(design)<-c("E0.5Hrs","E0Hrs","E2Hrs","E4Hrs","WsHrs") > fit<-lmFit(eset,design) > contrast.matrix<-makeContrasts(E0Hrs-WsHrs,E0.5Hrs-WsHrs,E2Hrs-WsHrs ,E4Hrs-WsHrs,levels=design) > fit2<-contrasts.fit(fit,contrast.matrix) > fit3<-eBayes(fit2) > tab<-topTableF(fit3,adjust="fdr",n=14010) the results from using topTableF give the following output example: ? ID?E0Hrs...WsHrs?E0.5Hrs...WsHrs?E2Hrs...WsHrs?E4Hrs...WsHrs?AveExpr?F ?P.Value?adj.P.Val 152598_at?-0.310260187?0.799814391?2.450974615?1.029883808?9.976246414 ?47.14172706?3.31E-08?0.00043327 148500_at?0.09207473?0.12454068?0.681084397?-0.01071873?3.732942206?42 .9457929?6.19E-08?0.00043327 147631_at?0.399408013?1.698760527?-0.067662327?-0.005311106?4.81034203 6?32.16015707?4.17E-07?0.001521181 154791_at?0.149794771?0.690132982?1.321826305?0.645885008?10.37671782? 31.9551017?4.34E-07?0.001521181 ? I would appreciate it if anyone could explain what the Pvalue column is measuring? Or is there something wrong with my codes as I'm thinking of seeing separate pvalue columns for each contrasts set above. ? Thanks Yemi

Hi Yemi, The p-value here is associated with the F-statistic, which is testing to see if *any* of your contrasts is significant. As I mentioned in my first response to your question, you probably want to look at the 'coef' argument of topTable, coef: column number or column name specifying which coefficient or contrast of the linear model is of interest. For 'topTable', can also be a vector of column subscripts, in which case the gene ranking is by F-statistic for that set of contrasts. which seems pretty clear to me. Best, Jim yemi yomi wrote: > Dear All, > > I carried out a time series analysis (4 time points with a common reference) analysis of my dataset using LIMMA method. I set the design and contrast matrix as: > > design<-model.matrix(~-1+factor(c(1,1,1,2,2,2,3,3,3,3,4,4,4,5,5,5,5, 5))) >> colnames(design)<-c("E0.5Hrs","E0Hrs","E2Hrs","E4Hrs","WsHrs") >> fit<-lmFit(eset,design) >> contrast.matrix<-makeContrasts(E0Hrs-WsHrs,E0.5Hrs-WsHrs,E2Hrs- WsHrs,E4Hrs-WsHrs,levels=design) >> fit2<-contrasts.fit(fit,contrast.matrix) >> fit3<-eBayes(fit2) >> tab<-topTableF(fit3,adjust="fdr",n=14010) > the results from using topTableF give the following output example: > > ID E0Hrs...WsHrs E0.5Hrs...WsHrs E2Hrs...WsHrs E4Hrs...WsHrs AveExpr F P.Value adj.P.Val > 152598_at -0.310260187 0.799814391 2.450974615 1.029883808 9.976246414 47.14172706 3.31E-08 0.00043327 > 148500_at 0.09207473 0.12454068 0.681084397 -0.01071873 3.732942206 42.9457929 6.19E-08 0.00043327 > 147631_at 0.399408013 1.698760527 -0.067662327 -0.005311106 4.810342036 32.16015707 4.17E-07 0.001521181 > 154791_at 0.149794771 0.690132982 1.321826305 0.645885008 10.37671782 31.9551017 4.34E-07 0.001521181 > > I would appreciate it if anyone could explain what the Pvalue column is measuring? Or is there something wrong with my codes as I'm thinking of seeing separate pvalue columns for each contrasts set above. > > Thanks > Yemi > > > > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- James W. MacDonald, M.S. Biostatistician Hildebrandt Lab 8220D MSRB III 1150 W. Medical Center Drive Ann Arbor MI 48109-5646 734-936-8662

Hi Yemi, The p-value here is associated with the F-statistic, which is testing to see if *any* of your contrasts is significant. As I mentioned in my first response to your question, you probably want to look at the 'coef' argument of topTable, coef: column number or column name specifying which coefficient or contrast of the linear model is of interest. For 'topTable', can also be a vector of column subscripts, in which case the gene ranking is by F-statistic for that set of contrasts. which seems pretty clear to me. Best, Jim yemi yomi wrote: > Dear All, > > I carried out a time series analysis (4 time points with a common reference) analysis of my dataset using LIMMA method. I set the design and contrast matrix as: > > design<-model.matrix(~-1+factor(c(1,1,1,2,2,2,3,3,3,3,4,4,4,5,5,5,5, 5))) >> colnames(design)<-c("E0.5Hrs","E0Hrs","E2Hrs","E4Hrs","WsHrs") >> fit<-lmFit(eset,design) >> contrast.matrix<-makeContrasts(E0Hrs-WsHrs,E0.5Hrs-WsHrs,E2Hrs- WsHrs,E4Hrs-WsHrs,levels=design) >> fit2<-contrasts.fit(fit,contrast.matrix) >> fit3<-eBayes(fit2) >> tab<-topTableF(fit3,adjust="fdr",n=14010) > the results from using topTableF give the following output example: > > ID E0Hrs...WsHrs E0.5Hrs...WsHrs E2Hrs...WsHrs E4Hrs...WsHrs AveExpr F P.Value adj.P.Val > 152598_at -0.310260187 0.799814391 2.450974615 1.029883808 9.976246414 47.14172706 3.31E-08 0.00043327 > 148500_at 0.09207473 0.12454068 0.681084397 -0.01071873 3.732942206 42.9457929 6.19E-08 0.00043327 > 147631_at 0.399408013 1.698760527 -0.067662327 -0.005311106 4.810342036 32.16015707 4.17E-07 0.001521181 > 154791_at 0.149794771 0.690132982 1.321826305 0.645885008 10.37671782 31.9551017 4.34E-07 0.001521181 > > I would appreciate it if anyone could explain what the Pvalue column is measuring? Or is there something wrong with my codes as I'm thinking of seeing separate pvalue columns for each contrasts set above. > > Thanks > Yemi > > > > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- James W. MacDonald, M.S. Biostatistician Hildebrandt Lab 8220D MSRB III 1150 W. Medical Center Drive Ann Arbor MI 48109-5646 734-936-8662