Hi I know this has been dealt with before, but I would really appreciate some help. I am anaylsing affy chips and I have three different sample types, lpl-high, lpl-low and brain. I construct a factor vector below TS <-c(rep(?lpl-high?,12),rep(?lpl-low?,10),rep(?brain?,17)) TS<- factor(TS, levels= c(?lpl-high?,?lpl-low?,?brain?)) I would simply want to make an analysis of variance, see wheter these three samples are differentially expressed (F value). I am not interested in any contrasts. How can I do that? Thanks a lot!!!!!

Create the contrasts. Use limma and FStat on the results. Or, to do a classical ANOVA, use apply to do an aov to each gene and use anova to get the F-stat from the aov output. e.g. myanova=function(x,y){ #x=expression for 1 gene #y=TS anova(aov(x~-1+y)) } ANOVAlist=apply(exprsmat,1,myanova,y=TS) Each component of ANOVAlist is the ANOVA table for 1 row of exprsmat. At 05:42 PM 1/10/2005 +0100, Auer Michael wrote: >Hi > >I know this has been dealt with before, but I would really appreciate some >help. >I am anaylsing affy chips and I have three different sample types, >lpl-high, lpl-low and brain. >I construct a factor vector below >TS <-c(rep("lpl-high",12),rep("lpl-low",10),rep("brain",17)) >TS<- factor(TS, levels= c("lpl-high","lpl-low","brain")) > >I would simply want to make an analysis of variance, see wheter these >three samples are differentially expressed (F value). I am not interested >in any contrasts. How can I do that? > >Thanks a lot!!!!! > >_______________________________________________ >Bioconductor mailing list >Bioconductor@stat.math.ethz.ch >https://stat.ethz.ch/mailman/listinfo/bioconductor Naomi S. Altman 814-865-3791 (voice) Associate Professor Bioinformatics Consulting Center Dept. of Statistics 814-863-7114 (fax) Penn State University 814-865-1348 (Statistics) University Park, PA 16802-2111

See limma User's Guide http://bioinf.wehi.edu.au/limma/usersguide.pdf middle of page 32. Gordon Mon Jan 10 17:42:20 CET 2005 >Hi > >I know this has been dealt with before, but I would really appreciate some >help. >I am anaylsing affy chips and I have three different sample types, >lpl-high, lpl-low and brain. >I construct a factor vector below >TS <-c(rep("lpl-high",12),rep("lpl-low",10),rep("brain",17)) >TS<- factor(TS, levels= c("lpl-high","lpl-low","brain")) > >I would simply want to make an analysis of variance, see wheter these >three samples are differentially expressed (F value). I am not interested >in any contrasts. How can I do that? > >Thanks a lot!!!!!

Is there a way to extract the original (without moderating the residual mean squares) F stat or F.p.value before apply eBayes step? I know I can do this by writing a function and apply it to each gene, but lmFit is much faster.... ...Tao ========================================================== Message: 11 Date: Tue, 11 Jan 2005 17:46:40 +1100 From: Gordon Smyth Subject: [BioC] limma design matrix To: "Auer Michael"michael.auer@meduniwien.ac.at Cc: bioconductor@stat.math.ethz.ch Message-ID: <6.2.0.14.1.20050111174221.0215e5d0@imaphost.wehi.edu.au> Content-Type: text/plain; charset="us-ascii"; format=flowed See limma User's Guide http://bioinf.wehi.edu.au/limma/usersguide.pdf middle of page 32. Gordon Mon Jan 10 17:42:20 CET 2005 >Hi > >I know this has been dealt with before, but I would really appreciate some >help. >I am anaylsing affy chips and I have three different sample types, >lpl-high, lpl-low and brain. >I construct a factor vector below >TS <-c(rep("lpl-high",12),rep("lpl-low",10),rep("brain",17)) >TS<- factor(TS, levels= c("lpl-high","lpl-low","brain")) > >I would simply want to make an analysis of variance, see wheter these >three samples are differentially expressed (F value). I am not interested >in any contrasts. How can I do that? > >Thanks a lot!!!!! __________________________________________________ [[alternative HTML version deleted]]