How to do limma on 3 factors with multiple levels (e.g 3-way anova) on microarray log values
0
Entering edit mode
Dan • 10
@Dan-24445
Last seen 17 days ago

I have microarray data to analyze with 3 factors: Genotype (wt, mut), Treatment (C, T), and Organ (s, r), each in 3 replicates. I found the guide for limma, I am just confused about how to "makeContrasts" to get all levels of main and interaction effect.

If this question is redundant, please provide the link, I could not find it.

Many thanks, Dan

Code should be placed in three backticks as shown below


# include your problematic code here with any corresponding output
# please also include the results of running the following in an R session

sessionInfo( )

interactions 3factors limma Microarray • 99 views
1
Entering edit mode
@gordon-smyth
Last seen just now
WEHI, Melbourne, Australia

As we try to explain in the limma documentation, testing for "main effect" and "interaction effect" is not a useful way to analyse a gene expression experiment. The above article explains a better way. It covers some of the same ground at the limma User's Guide but at greater length with more tutorial comments.

0
Entering edit mode

Thanks a lot!