Six sigma methodology and microarray power

["Phillip Staford" <biomining@hotmail.com>]

To reply to a poster last week who was searching for information about the 
six sigma methodology:
http://www.isixsigma.com/library/content/c000709.asp

is a premier source for information, history, and old/newly developed 
methods for the sixma process.

Hope that helps.  Also, does anyone have a good website for a descriptive 
breakdown of power analysis using pooled estimates of stdev?  I have a 
client who knows the power of a microarray chip (3 chips/1.3-fold change, 
takes into account the manufacturing variability) but he wants to determine 
the number of chips by taking into account the biological variability (each 
mouse responds differently to a treatment).  I suggested computing the 
standard deviation of a pilot experiment, 3 chips (technical replicates) and 
5 mice (biological replicates) for a total of 15 chips.  Is it correct to 
say that the standard deviation across all 15 chips would yield the 
biological variance, or would that be an overestimate (since the 
manufacturing variance is included)?  Should I recommend looking at stdev of 
only 1 chip per mouse (5 chips)?


Thanks in advance for any help;

Phillip Stafford

_________________________________________________________________
Protect your PC - get McAfee.com VirusScan Online 
http://clinic.mcafee.com/clinic/ibuy/campaign.asp?cid=3963