[BioC] LIMMA : trying to take contrast of non estimable coefficient

[Guillaume Meurice]

Dear James,

thanks for your quick answer !

> Your design matrix is not full rank. In other words, at least one column is a linear combination of other columns. This is why it is usually best to use functionality to create the design matrix, rather than trying to create one by yourself.

I'm trying to create the design matrix by myself, because in many of the project I have to analyse, there is a lot of parameters and "meta information" that came with the samples. And I have no idea (a priori) of how these meta information contribute to the observed intensity.

for example, for a sample, there is the treatment, the tissu where it came from, the observed effect of the treatment using methodology A, the observed effect of the treatment using methodology B, and so on.
Usually, I'm trying to have the design matrix that better fit the experimental design, to be sure to catch the specific effect of a given parameter.


> However, if you really want to create your own design matrix, you can always test it using functions in limma.
> 
> > is.fullrank(design)
> [1] FALSE
> > nonEstimable(design)
> [1] "RepClass.Unk" "SD"
> 
> So both your RepClass.Unk and SD columns are linear combinations of other existing columns. See also ?modelMatrix.

many thanks for all these information. I will check all these useful functionalities.


Best regards,
--
Guillaume