What Kinds Of Software Can Be Used For Multi-Platform Microarray Normalization For Meta-Analysis?

Question

[Cheng Zhongshan]

Dear members, I have two different microarray datasets from different platform, how can I normalize them before I carry out meta-analysis, for example, in this way, I can compare them directly? Thanks.

[Neilfws]

Is this not more or less the same question as When We Can Carry Out Microarray Meta-Analysis? ?

[Cheng Zhongshan]

Yes,a little bit similar, but I really don't know how to normalize all microarray together?

Answer 1

[User 59]

You have to understand that normalising, or attempting to normalise 2 independent experiments as one is fraught with issues, even if the same chip platform is being used for both experiments. It's hard enough making sure any within experiment batch effects are compensated for, let alone trying to do this for 2 separate experiments.

I think you need to understand that you are not going to be able to compare these two datasets directly in a straightforward 'lets just normalise it together' approach.

There are methods for cross-study normalisation, lots of them - whether you're combining gene expression measures across independent studies (Wang et al. or Stevens and Doerge) or combining other measures such as rank-ordering (as in RankProd as discussed) or p-values (Rhodes et al.). Then there are the Bayesian approaches (e.g. Conlon et al.).

Every method has caveats, issues and the more trivial the solution the more caveats. There is a BioConductor package called MADAM which implements some meta-analysis methods including the RankProducts approaches. There are also cross-study approaches implemented in the web-based analysis tool ArrayMining.net including empirical Bayes approaches (as in ComBat), median rank score normalisation, normalised discretization, quantile discretization - references to all of these are on the website.

Answer 2

[boczniak767]

This article reveals issues connected with metaanalysis.

Answer 3

[Chronos]

Regarding normalization - see also Empirical comparison of cross-platform normalization methods for gene expression data, which provides the CONOR R-package with 9 methods for cross-platform normalization.