[BioC] edgeR i get 377 significant genes where in DESeq i got 0

[Gordon K Smyth]

Dear Papori,

Probably you haven't made a mistake.  In our experience, this is the 
typical behaviour of the two packages.

The DESeq people should be applauded for trying something different, but 
commonsense would tell you that setting dispersions equal to the "maximum" 
of two extremes is likely to be conservative, especially when there are so 
few replicates.

Best wishes
Gordon

> Date: Mon, 12 Mar 2012 06:42:47 -0700 (PDT)
> From: "papori [guest]" <guest at bioconductor.org>
> To: bioconductor at r-project.org, drordh at gmail.com
> Subject: [BioC] edgeR i get 377 significant genes where in DESeq i got 0
>
> Hi,
> Assuming i have 2 files:
> 1's have 1,000,000 reads- one condition
> 2's have 3,000,000 reads- second condition
>
> i used rsem for differential expression for each of the 2 files.
>
> i used both programs to analyse differential expression.(edgeR & DESeq)
>
> i have a problem in the results..(maybe you can help me..)
> i used this manual http://manuals.bioinformatics.ucr.edu/home/ht-seq
> so may script is that:
>
> library("DESeq");
> conditions <- c("1","2")
> cds <- newCountDataSet(x,conditions)
> cds <- estimateSizeFactors(cds)
> cds <- estimateDispersions(cds,method="blind",sharingMode="maximum",fitType="local")
> res <- nbinomTest(cds,condA="1",condB="2")
> res[order(res$padj), ][1:4,]
>
> sigDESeq <- res[res$padj <= 0.05, ]
> sigDESeq <- na.omit(sigDESeq)
> sigDESeq <- as.character(sigDESeq[,1])
>
> library("edgeR");
> group <- factor(c(1,2))
> cdsR <- DGEList(counts=x, group=group)
> cdsR <- estimateCommonDisp(cdsR)
> cdsR <- estimateTagwiseDisp(cdsR)
> et <- exactTest(cdsR, pair=c("1", "2"))
>
> all <- as.data.frame(topTags(et, n=100000))
> sigedgeR <- all[all$adj.P.Val <= 0.05, ]
> sigedgeR <- na.omit(sigedgeR)
> sigedgeR <- row.names(sigedgeR)
>
> source("http://faculty.ucr.edu/~tgirke/Documents/R_BioCond/
> My_R_Scripts/overLapper.R")
> OLlist <- overLapper(setlist=list(DESeq=sigDESeq, edgeR=sigedgeR),
> sep="_", type="vennsets")
> counts <- sapply(OLlist$Venn_List, length)
> vennPlot(counts=counts)
> counts
>
> oveLapp <- OLlist$Venn_List
>
> $$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$
>
> i want to check Differential expression between biological replicates.
> (here: i have only 2)
> 1 & 2 are biological replicates.
>
> the problem is:
> in edgeR i get 377 significant genes where in DESeq i got 0 (ZERO)
>
> it looks weird to me, but i do'nt know what i did wrong..
>
> Do you know?
>
>
> Thanks,
> Pap
>
> -- output of sessionInfo():
>
> R version 2.14.0 (2011-10-31)
> Platform: x86_64-unknown-linux-gnu (64-bit)
>
> locale:
> [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C               LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8
> [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8    LC_PAPER=C                 LC_NAME=C
> [9] LC_ADDRESS=C               LC_TELEPHONE=C             LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
>
> attached base packages:
> [1] stats     graphics  grDevices utils     datasets  methods   base
>
> other attached packages:
> [1] edgeR_2.4.3  limma_3.10.2
>
> loaded via a namespace (and not attached):
> [1] annotate_1.32.1       AnnotationDbi_1.16.16 Biobase_2.14.0        DBI_0.2-5             DESeq_1.6.1
> [6] genefilter_1.36.0     geneplotter_1.32.1    grid_2.14.0           IRanges_1.12.6        RColorBrewer_1.0-5
> [11] RSQLite_0.11.1        splines_2.14.0        survival_2.36-12      tools_2.14.0          xtable_1.7-0
>
> --
> Sent via the guest posting facility at bioconductor.org.
>

______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}