Hi all!!

I am making a differential expression analysis of microarray data with two contrasts. My design matrix is like this

  A_1  A_2    B
   1      0        0
   1      0        0
   1      0        0
   0      1        0
   0      1        0
   0      1        0
   0      1        0
   0      0        1
   0      0        1
   0      0        1
   0      0        1
   0      0        1
   0      0        1
   0      0        1

And my contrast matrix, made with the makeContrast function is this

       Contrasts
Levels  A_1-B   A_2 - B
   A_1     1       0
   A_2     0       1
   B      -1      -1

I want to find the GO_terms associated to both contrasts, for that I am using the camera function like this

camera(Exprs, Index, design, contrast = contMatrix) 

Where:

Exprs is my matrix of log-expression values 

Index is the list of GO terms associated to the genes in the Exprs matrix

design is the design matrix I showed you before 

and as contrast I input my contMatrix object i showed you before.

My doubt is if camera output is actually the GO_terms taking account the TWO contrasts defined. 

If not, how can I get that result.

Thanks!!

camera() only uses one contrast. I wonder what you expected it to do with two contrasts? Did you expect a test by F-statistic? Or did you expect two separate analyses?

When in doubt it's not a bad idea to read the documentation. The help page for camera() says that the contrast argument is:

contrast of the linear model coefficients for which the test is required. Can be an integer specifying a column of design, or else a numeric vector of same length as the number of columns of design.

As you can see, the documentation says that the contrast can be either an integer or a vector. It cannot be a matrix.