[BioC] experiment-design-advice-request

[Naomi Altman]

Design 1 is better than 2 but is unbalanced for the dye effect.  If there 
could be a dye by gene interaction you may have some difficulties in 
interpreting the design.

Design 2 is unbalanced for the main effects.  You would be better to 
complete the loop, with 4 hybridizations.

I am not sure what you mean by "dyeswap and 2 replicates".  Are you using 6 
or 3 arrays, or is this the base design which you intend to replicate?  If 
you plan to have 8 or 12 arrays, you have many more choices for the design.

Finally, technical replication is not that informative.  If at all 
possible, each hybridization should be done with an independent biological 
sample.  Technical replication is done mainly if the cost of creating an 
independent biological sample is too high.  (Technical replication is also 
done due to the misconception that it is necessary for dye-swapping.)

Just a note - statisticians consider the control to be a condition, so you 
actually have 2 conditions.

--Naomi

At 02:49 PM 4/30/2005, you wrote:
>hi friends
>we are designing a microarray experiment, where there are two different 
>mouse strains (A,B)...
>and one condition (Peptide - P and Saline - S). we expect the mouse 
>strains to express differentially even under normal  (saline) 
>conditions...so we did not want to go for pooling the controls, to have a 
>common - pooled control... under this scenario...
>
>which of the following designs would you suggest....(we have planned to 
>use dyeswap and 2 replicates for each hybrisation)....
>
>1:
>
>AP---BP
>|  \  /   |
>|  /  \   |
>AS---BS
>
>(all pairs - with 6 hybridisations)
>       OR
>
>2:
>
>AP
>|
>AS---BS
>          |
>        BP
>
>(with just 3 hybridisations, so that we could deduce AP-BP,using AP-AS-BS 
>and rest like that...)
>
>i request your valuable advice in this regard....
>thanks
>
>vijay
>graduate student
>department of biological sciences
>University of Southern Mississippi
>MS
>
>
>__________________________________________________
>
>
>
>         [[alternative HTML version deleted]]
>
>_______________________________________________
>Bioconductor mailing list
>Bioconductor at stat.math.ethz.ch
>https://stat.ethz.ch/mailman/listinfo/bioconductor

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

	[[alternative HTML version deleted]]