[BioC] Query regarding limma

[Maciej Jończyk]

Hi Roopa,

> results <- decideTests(fit2, adjust="fdr", p=0.05)
> summary(results)
> write.table(results,file="myresults.txt")
>
> The results table shows ~10,000 genes to be upregulated and ~12,000 
> genes
> to be down regulated.
>
> My question is how can I get fold change values associated with these 
> genes
> in the results file?

You can create object with needed columns binded:

e.g.

x=cbind(fit$coefficients,fit$p.value,p.adjust(fit$p.value,"BH"),separate)

Where "fit" is the result of lmFit and then eBayes commands - it 
contains
"coefficients" column with (mean) log2 fold change for each gene.

*In fact it haven't been clear for me at first - but I compared it to 
the output of topTable (logFC column) and it is equal.*

Back to the table - proposed example will give you:
log2 fold change,
raw p-value,
corrected p-value (here Benjamini-Hochberg, equall to "fdr"),
and finally change direction for significant genes (where separate is 
result of decideTests)

Before "write.table" you can change column names with 
colnames(x)=c(...)


HTH

-- 
Maciej Jonczyk,
Department of Plant Molecular Ecophysiology
Faculty of Biology, University of Warsaw
02-096 Warsaw, Miecznikowa 1
Poland


-- 
This email was Anti Virus checked by Astaro Security Gateway. http://www.astaro.com