This is my first time working with microarray analysis and I have a few very basic questions.

I am planning on using a custom array design from someone else that is produced by Agilent.

I am planning on performing 3 microarrays on Control, Treatment 1, and Treatment 2 samples. My goal is to find the differential expressed (particularly upregulated genes) in each treatment. Then I'd like to statistically compare them to one another. I believe I should do this as a two-color array.

(1) When considering biological replicates, should the minimum be that I take 3 different RNA sample extractions and run 3 microarrays for each treatment? That would mean I need to buy 9 chips. And perform 9 RNA extractions all from a different group of individuals 'pooling' (I'm using mosquitoes).

I ask this question so I can better understand the cost of this experiment set up.

(2) reference versus loop design

So reference design is my 'control - untreated' and I compare each treatment to the control directly, and to other treatments indirectly

In loop design, I could compare all of them to the control directly and to each other directly (so to speak)

I have not chosen which one I want to use yet, but I wanted to give a better understanding of what I am here to accomplish.

(3) I've brought together materials from relevant papers and I think I have that all situated. Does this look like the proper, very brief method flow:

i. RNA extraction

ii. RNA isolation (include extra removal from genomic DNA)

iii. RNA amplification

iv. cDNA synthesis and Labeling

v. amplify cDNA

vi. hybridize

vii. wash

viii. scan

I would appreciate any pointers for major flaws I may have proposed. And greatly appreciate all help in advance.