[BioC] Modelling interaction using limma package

[Gordon Smyth]

At 03:41 AM 15/09/2004, Fangxin Hong wrote:
>Does anyone know how to model interaction (two-way ANOVA) using limma
>package (lmFit)?

There is an essay on this topic in the section "Special Designs" in the new 
Limma User's Guide, available from 
http://bioinf.wehi.edu.au/limma/usersguide.pdf. If you wait a few days, 
this new material will be in the Bioconductor version of limma.

>Also, it seems that eBayes can be applied to other model fit object, like
>'lm.series', 'gls.series' and  'rlm.series' beside 'lmFit'. Does anyone
>use this before?

lmFit() is just a front end to the other functions that you mention. All 
the functions fit linear models in a unified way. I recommend that you 
stick with lmFit().

Gordon

>Thanks a lot!  I am quite new to bioconductor...
>
>Fangxin
>
>
>--
>Fangxin Hong, Ph.D.
>Bioinformatics Specialist
>Plant Biology Laboratory
>The Salk Institute
>10010 N. Torrey Pines Rd.
>La Jolla, CA 92037
>E-mail: fhong at salk.edu