[BioC] Merging microarray datasets

pingzhao Hu phu at sickkids.ca
Thu Apr 24 16:30:20 CEST 2008

To answer your questions, you may need to know which stages you are 
trying to integrate your data sets.
Roughly speaking, data sets can be integrated at three stages:
Say you have two data sets A (affy) and B (cDNA)

(1) early stage: first normalize data sets A and B, separately; then 
somehow rescale normalized A and B; finally put them together
(2) middle level stage; first normalize data sets A and B, then 
calculate effect size (or like meansures/methods) for each gene in A 
and B, separately;
      finally you can use random/fixed effects model to integrate the 
effect sizes (like Choi et al. 2003 method)
(3) late stage: first calculate normalize data sets A and B, 
separately; calculate p-value/test statistic for each gene in each 
study (standard methods used
     in single study); finally combine the p-values (say using Fisher 

There are some bioconductor packages can do (2) and (3) -like RankProd.


Pingzhao Hu
Statistical Analysis Facility
The Centre for Applied Genomics (TCAG)
The Hospital for Sick Children Research Institute
MaRS Centre - East Tower
101 College Street, Room 15-705
Toronto, Ontario, M5G 1L7, Canada
Tel.: (416) 813-7654 x6016
Email: phu at sickkids.ca
Web: http://www.tcag.ca/statisticalAnalysis.html

More information about the Bioconductor mailing list