[BioC] heat map for a time course study

[Thomas H. Hampton]

Andrew,

Try using heatmap.2  (in library gplots) to make a picture of the 100 genes with the lowest p value in whatever test you are using to assess significance. 

Something along these lines:

heatmap.2 (exprs(RMA[<YOUR GENES>,]))

where <YOUR GENES> is list of the 100 genes with smallest p values. 

Best,

Tom




On Oct 10, 2011, at 1:06 PM, blockaa at huskers.unl.edu wrote:

> Hello BioConductor world,
> 
> 
> I am doing a time course study using Affymetrix microarrays. I have three time points (0, 4, and 6 hours) and a control and treatment. There is no treatment at 0 hours because the genes would be the same. I am able to get the significant transcripts, but I can not create the heat maps I want. I have searched the literature and the internet and have not found what I was looking for.  I would like two heat maps: one for the controls versus treatments for the different time points (two columns or groups, 4 and 6 hr) and the second heat map comparing the different times (three columns or groups, 0 to 4 and 6 hr and 4 to 6 hr). What would be the best way to do this? Is there a way to combine the the results from the different microarrays in the same control or treatment to form one column instead of the multiple columns? Looking at other heat maps, I could not find an example of the combined columns.   I have about 200 different transcripts for the the controls versus treatments !
> for the different time. Is there a way to select the number of genes in the heat map.  I may be think about the heat maps wrong, so if I am what is the best way to think about heat maps and the code or instructions, for a new person, for the right type of heat map.
> 
> 
> Here is my code (control time [c0, c4, c6] or treatment time[t4,t6]):
> 
> 
> library(affy)
> 
> library(limma)
> 
> 
> targets = readTargets("time course.txt")
> 
> raw = ReadAffy(filenames=targets$FileName)
> 
> data = read.AnnotatedDataFrame("time course.txt", sep="", fill=TRUE)
> 
> phenoData(raw) = data[sampleNames(raw),]
> 
> RMA = rma(raw)
> 
> 
> lev = c("c0","c4","t4","c6","t6")
> 
> f = factor(data$Target, levels=lev)
> 
> design = model.matrix(~0+f)
> 
> colnames(design) = lev
> 
> fit = lmFit(RMA, design)
> 
> cont.wt = makeContrasts(
> 
>    "t4-c4",
> 
>    "t6-c6",
> 
>    c0t4 = c0-(t4-c4),
> 
>    c0t6 = c0-(t6-c6),
> 
>    t4t6 = (t4-c4)-(t6-c6),
> 
>    levels=design)
> 
> fit2 = contrasts.fit(fit, cont.wt)
> 
> fit2 = eBayes(fit2)
> 
> 
> results = classifyTestsF(fit2, p.value=0.05)
> 
> summary(results)
> 
> 
> Thank you in advance.
> 
> Andrew Block
> Graduate Student
> Nebraska Center of Virology
> University of Nebraska at Lincoln
> 
> 	[[alternative HTML version deleted]]
> 
> _______________________________________________
> Bioconductor mailing list
> Bioconductor at r-project.org
> https://stat.ethz.ch/mailman/listinfo/bioconductor
> Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor
>