Interpreting topTable results for limma factorial design
1
0
Entering edit mode
Sally ▴ 250
@sally-2430
Last seen 9.6 years ago
I have a 2x2 factorial design which I ran through limma. The factors are (1) species [(Coho (c)) and Sockeye (s))] and (2) sample time (0, 24, 48 and 96 hours). The R script used was: source("http://bioconductor.org/biocLite.R") library(limma) library(Biobase) exprdata<-read.table("exprsData.txt", header=TRUE,sep="\t",row.names=1,as.is=TRUE,fill=TRUE,) phenotypicdata<-read.table("phenotypicdata.txt",row.names=1,header=TRU E,sep="\t") myexprdata<-as.matrix(exprdata) myphenotypicdata<-as.data.frame(phenotypicdata) adf<-new("AnnotatedDataFrame",data=phenotypicdata) eset<-new("ExpressionSet",exprs=myexprdata,phenoData=adf) targets <- readTargets("targets.txt") TS <- paste(targets$Species, targets$Time, sep=".") TS <- factor(TS) design <- model.matrix(~0+TS) colnames(design) <- levels(TS) fit <- lmFit(eset, design) cont.matrix<-makeContrasts(s0vss24=s.0-s.24, s24vss48=s.24-s.48, s48vss96=s.48-s.96, c0vsc24=c.0-c.24, c24vsc48=c.24-c.48, c48vsc96=c.48-c.96, s0vsc0=s.0-c.0, s24vsc24=s.24-c.24, s48vsc48=s.48-c.48, s96vsc96=s.96-c.96, levels=design) fit2 <- contrasts.fit(fit, cont.matrix) fit2 <- eBayes(fit2) c48vsc96<-topTable(fit2,coef="c48vsc96",number=400,adjust.method="BH", p.value=1) I have appended an Excel file which includes both the topTable results and the average M values for both coho 48 hours and coho 96 hours for those genes with a p value less than 0.05. My questions are: For gene #1, what does the logFC mean? At which sample time (48 or 96 hours) is this gene down-regulated -2.67 (fold change)? The logFC does not appear to correlate with either of the two M values (48hrs = +0.63, 96 hrs = +2.42)? Something seems wrong to me. Thanks Sally Goldes
limma limma • 1.1k views
ADD COMMENT
0
Entering edit mode
@james-w-macdonald-5106
Last seen 2 hours ago
United States
Hi Sally, You aren't going to be able to get an Excel attachment through the listserv (it will be stripped). Your best bet is to put the data inline. Best, Jim Sally wrote: > I have a 2x2 factorial design which I ran through limma. The factors are (1) species [(Coho (c)) and Sockeye (s))] and (2) sample time (0, 24, 48 and 96 hours). > > > > The R script used was: > > > > source("http://bioconductor.org/biocLite.R") > > library(limma) > > library(Biobase) > > exprdata<-read.table("exprsData.txt", header=TRUE,sep="\t",row.names=1,as.is=TRUE,fill=TRUE,) > > phenotypicdata<-read.table("phenotypicdata.txt",row.names=1,header=T RUE,sep="\t") > > myexprdata<-as.matrix(exprdata) > > myphenotypicdata<-as.data.frame(phenotypicdata) > > adf<-new("AnnotatedDataFrame",data=phenotypicdata) > > eset<-new("ExpressionSet",exprs=myexprdata,phenoData=adf) > > targets <- readTargets("targets.txt") > > TS <- paste(targets$Species, targets$Time, sep=".") > > TS <- factor(TS) > > design <- model.matrix(~0+TS) > > colnames(design) <- levels(TS) > > fit <- lmFit(eset, design) > > cont.matrix<-makeContrasts(s0vss24=s.0-s.24, s24vss48=s.24-s.48, s48vss96=s.48-s.96, c0vsc24=c.0-c.24, c24vsc48=c.24-c.48, c48vsc96=c.48-c.96, s0vsc0=s.0-c.0, s24vsc24=s.24-c.24, s48vsc48=s.48-c.48, s96vsc96=s.96-c.96, levels=design) > > fit2 <- contrasts.fit(fit, cont.matrix) > > fit2 <- eBayes(fit2) > > c48vsc96<-topTable(fit2,coef="c48vsc96",number=400,adjust.method="BH ",p.value=1) > > > > I have appended an Excel file which includes both the topTable results and the average M values for both coho 48 hours and coho 96 hours for those genes with a p value less than 0.05. > > > > My questions are: > > > > For gene #1, what does the logFC mean? > > At which sample time (48 or 96 hours) is this gene down-regulated -2.67 (fold change)? > > The logFC does not appear to correlate with either of the two M values (48hrs = +0.63, 96 hrs = +2.42)? > > Something seems wrong to me. > > > > Thanks > > > > Sally Goldes > > > > > > > -------------------------------------------------------------------- ---- > > _______________________________________________ > 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 Affymetrix and cDNA Microarray Core University of Michigan Cancer Center 1500 E. Medical Center Drive 7410 CCGC Ann Arbor MI 48109 734-647-5623
ADD COMMENT

Login before adding your answer.

Traffic: 886 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6