DESeqDataSetFromMatrix giving an error

Question

[natheshka]

Hi,

I am trying to follow the example in MLSeq documentation. But I keep on getting errors. The latest one I am getting is

Error in DESeqDataSet(se, design = design, ignoreRank): counts matrix should be numeric, currently it has mode: logical

My code is

data.trainS4 = DESeqDataSetFromMatrix(countData = data.train, colData = classtr,
design = formula(~condition))

I am really new to R and learning it as I am following these examples. Any help would be really great.

Thanks..

[swbarnes2]

Well, what does data.train look like? Can you make a subset file of the first few lines and get that to import?

[ATpoint]

The output of head(data.train)would do.

Answer 1

[c.chakraborty]

Check if your matrix is raw read counts. If it is normalized to the library, i.e counts would look like - 38.713, 410.006, then DESeqDataSetFromMatrix gives an error. DESeq2 is designed for raw read counts.

Answer 2

[Quetzal]

DESeq2 just works on numeric data . If you don't have a matrix with numeric data DESeq2 gives an error .In that case you should use limma not DESeq2.