[BioC] Merging microarray datasets
pingzhao Hu
phu at sickkids.ca
Thu Apr 24 16:30:20 CEST 2008
Kathy,
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
method)/test-statistic.
There are some bioconductor packages can do (2) and (3) -like RankProd.
Pingzhao
========================================
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