[BioC] Two color Array 8 hybridization Group Design Contrast Question

Naomi Altman naomi at stat.psu.edu
Fri Nov 10 19:28:18 CET 2006

Dear Seungmin,

You are at a university with one of the top departments of Statistics 
and Biostatistics.  Probably there is a Statistical Consulting 
Center.  Even if the SCC consultants do not know limma or 
Bioconductor, they can help set up the model and the contrasts.  (Of 
course, Berkeley is the Ph.D or professional home of many 
Bioconductor contributors.)

However, if all you want is to select genes that differentially 
express under at least one condition, you could use 1-way ANOVA, 
using the treatment names you have displayed below.  I do not have 
the manual in front of me, but I am sure it is one of the 
examples.  You do not need to fit this as a factorial 
experiment.  You can use the ANOVA F-test as recently discussed on 
this forum, or you can use the individual t-tests.

--Naomi Altman

At 01:29 PM 11/9/2006, lee wrote:
>   I have mouse 8 hybridization Groups.
>   2 tissues (Ent, Liv)
>   2 ages (4wk, 8wk)
>   2 different diets. (IOL,ID) compared to control diet
>   2 different mutation.(sla, HFE) compared to wild type.
>   The following is the names for each hybridization group.
>   "sla4wkEnt","HFE4wkEnt",
>   "HFE4wkLiv","sla4wkLiv","IOL4wkLiv",
>   "HFE8wkLiv","ID8wkLiv","IOL8wkLiv"
>   I have at least 4 hybridizations for each hybridization group. 
> Most of them have around 6 ~12 hybridizations.
>   Each hybridization represents individual biological samples(mouse).
>   I have switched dyes to label control or experimental samples.
>   For example, 3 of experimental mouse samples were labeled with 
> Cy5 and 3 of them are labeled with Cy3.
>   I studied LIMMA program and was able to read all gpr files and 
> normalize the data.
>   After this, I don't know how to design and make contrast to know 
> which genes are significantly different at least one group among 8 
> hybridization groups.
>   After finding out the candidate genes, I want do cluster analysis 
> to find out which hybridization groups are similar in terms of gene 
> expression profiling and which gene groups behave similarly across 
> different hybridization groups.
>   I studied User's Guide a lot. But most of them are for Affy data. 
> After trying copying and modifying examples on my own, I find out 
> it is out of my ability to do analysis for my data. Could you help 
> me how to do this kind of analysis or give me detailed examples 
> that I can use with modification for my data?
>   Thank you so much for your help in advance.
>   Seungmin
>Seung-Min Lee
>graduate student
>244 Morgan Hall
>Molecular&Biochemical Nutrition
>University of California at Berkeley
>lab phone (510)643-2351
>lab fax(510)642-0535
>         [[alternative HTML version deleted]]
>Bioconductor mailing list
>Bioconductor at stat.math.ethz.ch
>Search the archives: 

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

More information about the Bioconductor mailing list