Integrative Analysis of "-Omics" Data Using Penalty Functions.

Qing Zhao,Xingjie Shi,Jian Huang,Jin Liu,Yang Li,Shuangge Ma
DOI: https://doi.org/10.1002/wics.1322
2015-01-01
Abstract:In the analysis of omics data, integrative analysis provides an effective way of pooling information across multiple datasets or multiple correlated responses, and can be more effective than single-dataset (response) analysis. Multiple families of integrative analysis methods have been proposed in the literature. The current review focuses on the penalization methods. Special attention is paid to sparse meta-analysis methods that pool summary statistics across datasets, and integrative analysis methods that pool raw data across datasets. We discuss their formulation and rationale. Beyond "standard" penalized selection, we also review contrasted penalization and Laplacian penalization which accommodate finer data structures. The computational aspects, including computational algorithms and tuning parameter selection, are examined. This review concludes with possible limitations and extensions.
What problem does this paper attempt to address?