[BioC] model3 (mixed-effect) ANOVA in any bioc package?

[Naomi Altman]

You could start by looking at limma and maanova.  In limma, you would 
treat patients as blocks.  Limma allows only one random effect.  I 
have not used maanova, but I believe it allows multiple random effects.

--Naomi

At 05:56 AM 8/10/2009, Riccardo Rossi wrote:
>Hello,
>
>Is there a specific packafe or function to perform model 3 (mixed 
>effects) ANOVAs on expression data? Is anyone able to address me to 
>the correct bioc package/function if any? (I was not able so far.)
>
>Here's the details of what I have and what I need to do:
>1) the data
>57 genelists, one list for each sample I have
>These 57 genelists belongs to 17 different sample type, let's say 
>cell populations
>The 57 samples come from 30 different experiments (or sorting/purifications)
>Basically is like having 57 samples grouped in 17 cell types 
>obtained from 30 different patients, where "cell types" is a fixed 
>factor and "patients" is a random factor (infact samples belonging 
>to each cell population come from different patients/sorting, 
>according to availability of the moment)
>2) what I need to do
>I would like to pick genes significantly differentially expressed 
>among the 17 samples, considering also the "contribute" of the 
>random factor (patients, i.e. sorting) to sample variability. So far 
>I only made 1way (model1) ANOVAs considering the cell populations 
>grouping (fixed effect), I need to make a Model3 (mixed effect) 
>ANOVA considering also the sorting/purification grouping as a random effect.
>
>Thanks to everyone!,
>Ric
>
>
>
>----------------------------------------
>Riccardo L. Rossi, PhD
>Integrative Biology Program
>Fondazione INGM
>Via F. Sforza 28
>20122 Milano, IT
>phone: +39 02 00662326
>fax: +39 02 00662346
>rossi at ingm.it
>www.ingm.it
>SOSTIENI LA RICERCA: devolvi il 5 per mille delle imposte alla 
>FONDAZIONE INGM Istituto Nazionale Genetica Molecolare di Milano. 
>Devi solo inserire il codice fiscale della Fondazione - 04175700964 
>- in tutti i modelli CUD, 730 e UNICO nella sezione relativa al 
>finanziamento per la ricerca scientifica.
>
>         [[alternative HTML version deleted]]
>
>_______________________________________________
>Bioconductor mailing list
>Bioconductor at stat.math.ethz.ch
>https://stat.ethz.ch/mailman/listinfo/bioconductor
>Search the archives: 
>http://news.gmane.org/gmane.science.biology.informatics.conductor

Naomi S. Altman                                814-865-3791 (voice)
Associate Professor
Dept. of Statistics                              814-863-7114 (fax)
Penn State University                         814-865-1348 (Statistics)
University Park, PA 16802-2111