[BioC] EdgeR annotation
Gordon K Smyth
smyth at wehi.EDU.AU
Mon Aug 26 06:50:09 CEST 2013
Dear Robin,
I don't think the problem is with edgeR. You are setting
genes=data[,0:1]
However there is no column 0, so 0:1 reduces to column 1, which is being
set as annotation. But you also input column 1 as a count:
counts=data[,1:18]
These can't both be right.
I suspect you just want:
y <- DGEList(counts=data)
Best wishes
Gordon
----- original message ------
[R] EdgeR annotation
Robin Mjelle robinmjelle at gmail.com
Sat Aug 24 13:50:55 CEST 2013
after updating R and edgeR I lost the annotations in the final
Diff.Expressed matrix (toptags) when running the edgeR pipeline. How do I
get the row.names from the data matrix into the topTag-matrix?
data <- read.table("KO_and_WT_Summary_miRNA_Expression.csv", row.names=1,
sep="", header=T)
keep <- rowSums(cpm(data)>2) >=2
data <- data[keep, ]
table(keep)
y <- DGEList(counts=data[,1:18], genes=data[,0:1])
y <- calcNormFactors(y)
y$samples
plotMDS(y,main="")
Time=c("0.25h","0.5h","1h","2h","3h","6h","12h","24h","48h","0.25h","0.5h","1h","2h","3h","6h","12h","24h","48h")
Condition=c("KO","KO","KO","KO","KO","KO","KO","KO","KO","WT","WT","WT","WT","WT","WT","WT","WT","WT")
design <- model.matrix(~0+Time+Condition)
rownames(design) <- colnames(y)
y <- estimateGLMCommonDisp(y, design, verbose=TRUE,
method="deviance",robust=TRUE, subset=NULL)
y <- estimateGLMTrendedDisp(y, design)
y <- estimateGLMTagwiseDisp(y, design)
fit <- glmFit(y, design)
lrt <- glmLRT(fit)
topTags(lrt)
Coefficient: ConditionWT
genes logFC logCPM LR PValue FDR
189 5128 -11.028422 7.905804 4456.297 0 0
188 12271 -10.582267 9.061326 5232.075 0 0
167 121120 -9.831894 12.475576 5957.104 0 0
34 255235 -9.771266 13.592968 7355.592 0 0
168 311906 -9.597952 13.907951 10710.111 0 0
166 631262 -9.592550 14.932018 11719.222 0 0
79 79 9.517226 11.466696 7964.269 0 0
169 2512 -8.946429 6.758584 2502.548 0 0
448 3711 -7.650068 7.764682 2914.784 0 0
32 260769 -7.412197 13.633352 4906.198 0 0
______________________________________________________________________
The information in this email is confidential and intend...{{dropped:4}}
More information about the Bioconductor
mailing list