Hi

We have a big patient cohort and are wondering on the best ways to integrate the data. So far we have been using WGCNA on the microarray data with moderate success, but the question is how to we approach this when multiple genomic platforms must all be integrated? So we are interested in clustering the data using the platforms together with respect to clinical data (consensus clustering, iclusterplus), then also prediction of prognosis, and also network construction/ module generation with correlations to clinical data (WGCNA, ARACNE). Does anyone have any advice here? Any neat packages I am missing?

Kind thanks,

Chris

Hi Chris,

the section "Tools for integrative genomic data analysis" and Table 4 in http://www.ncbi.nlm.nih.gov/pubmed/26463000 survey some more tools you might be interested in. 

Also, we have a working prototype of a data class for data management of this kind of study (https://github.com/vjcitn/biocMultiAssay), built vignette at http://rpubs.com/lwaldron/MultiAssayExperiment. It doesn't implement high-level methods like the ones you mention, but rather helps make sure your samples are aligned correctly and helps to do things like subset or intersect by genes or whatever features.  If you want to try it out, I'd be happy to help. 

Best,

Levi