[BioC] Help for setting up design and contrast matrix

John Antony Gaspar gasparj at uni-koeln.de
Tue Dec 7 10:18:23 CET 2010


Dear BioC,

I would greatly appreciate your help if you could help me out in setting 
up design and contrast matrix for the following experimental set up;

I have 5 groups of samples with their WT (ES) type, Differentiated type 
and Differentiated_purified( with  puromycin treatment on differentiated 
to have pure cells)type.
To make it clear aES, aDiff, aDiff_pure, bES, bDiff, bDiff_pure, cES, 
aDiff, cDiff_pure,
dES, dDiff, dDiff_pure, eES, eDiff, eDiff_pure. All are with biological 
triplicates.

I wish to model interaction (two way anova) using limma package's lmFit 
function (differentiation and pure_treatment)  to find out the effect of 
Differentiation and Differentiated_purification.
With regard to differentiation, I would try with setting the level;
 
Diff<-factor(c(1,1,1,2,2,2,2,2,2,1,1,1,2,2,2,2,2,2,1,1,1,2,2,2,2,2,2,1,1,1,2,2,2,2,2,2,1,1,1,2,2,2,2,2,2)) 
- Here all WT as 1 and other as 2. Is it right?.

With regard to differentiated_purification,
Diff_pure<-factor(c(1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,2,2,2,1,1,1,1,1,1,2,2,2)) 
- WT and Diff as 1 and Diff_purified as 2.


design<-model.matrix(~Diff+Diff_pure) Is it right?. and how to set the 
model if I wish to have the interaction effect between Diff and Diff_pure?.

Please help me out in setting up the Contrast matrix too;
Is it right?
contrast.matrix <- makeContrasts( ,level=design)).

I am in need of your  valuable and timely help , please help me out.

Thanking you in advance,
antony



More information about the Bioconductor mailing list