|
Abstract:
|
Most integrative studies using multiple 'omic data types rely on dimension reduction techniques for each platform before attempting combined analysis, resulting in information and power loss. Though non-reductive testing approaches have become more common, many are restricted to candidate gene interrogations, and do not reflect the highly likely network-level interactions between genes. An efficient screening approach leveraging multi-omic data across gene sets is critical for hypothesis generation. We propose an efficient variance-component based screening approach across multi-platform genomic data on the level of whole gene sets. Our methods are applicable to various disease models regardless whether the underlying true model is known (iTEGS) or unknown (iNOTE). In real data, we identified a total of 28 gene sets with significant joint epigenomic and transcriptomic effects on one-year lung cancer survival. The testing procedure for the gene set is self-contained, and can easily be extended to include more or different genetic platforms. iTEGS and iNOTE implemented in R are freely available through the inote package at https://cran.r-project.org//.
|
