Integrating Multiple Data Sources with Interactions in Multi-Omics Using Cooperative Learning

TL;DR

This paper introduces an interaction model that combines pliable lasso and cooperative learning to effectively integrate multiple multi-omics data sources, capturing interactions for improved prediction and variable selection.

Contribution

The paper presents a novel integrated modeling approach that incorporates interactions across multiple data sources in multi-omics, enhancing prediction and feature selection capabilities.

Findings

Effective in modeling multi-source data with interactions

Improves prediction performance in real datasets

Enhances variable selection accuracy

Abstract

Modeling with multi-omics data presents multiple challenges such as the high-dimensionality of the problem ($p \gg n$), the presence of interactions between features, and the need for integration between multiple data sources. We establish an interaction model that allows for the inclusion of multiple sources of data from the integration of two existing methods, pliable lasso and cooperative learning. The integrated model is tested both on simulation studies and on real multi-omics datasets for predicting labor onset and cancer treatment response. The results show that the model is effective in modeling multi-source data in various scenarios where interactions are present, both in terms of prediction performance and selection of relevant variables.