Hi all,

I am using DMRcate to find differentially methylated regions in illumina 450k data between one group of 4 samples and another group of 9 samples.

I followed all the steps of DMRcate user's guide and I build the design matrix in a similar manner,

and this is it:

str(design) num [1:13, 1:14] 1 1 1 1 1 1 1 1 1 1 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:13] "1" "2" "3" "4" ... ..$ : chr [1:14] "(Intercept)" "patient020" "patient021" "patient023" ... - attr(*, "assign")= int [1:14] 0 1 1 1 1 1 1 1 1 1 ... - attr(*, "contrasts")=List of 2 ..$ patient: chr "contr.treatment" ..$ type : chr "contr.treatment"

But when I call the cpg.annotate(), I have this error:

myannotation <- cpg.annotate(myMvalue, analysis.type = "differential", design = design, coef = 14) Coefficients not estimable: typeGroup2 Error in ebayes(fit = fit, proportion = proportion, stdev.coef.lim = stdev.coef.lim, : No residual degrees of freedom in linear model fits Inoltre: Warning message: Partial NA coefficients for 425827 probe(s)

In the DMRcate example dataset, that I have tested, the differential analysis is paired 38 vs 38

and I think that the error probably is in the design matrix building of my dataset.

So how to set the design matrix correctly?

Thanks in advance, Regards

Giovanni

Hi Tim,

Many Thanks, It works fine!

In particular I had already tried to build the matrix by

used to calculate DMPs, but I wrong in the coef argument in cpg.annotate call

having again error. Maybe I misinterpreted the cpg.annotate call in the paired analysis of example dataset.

Then using the coef argument

that reflects my originary

I got DMRs and this is the design matrix:

Good luck,

Giovanni