Multi-omic Data Integration and Feature Selection for Survival-Based Patient Stratification via Supervised Concrete Autoencoders

Pedro Henrique da Costa Avelar, Roman Laddach, Sophia Karagiannis, Min Wu, Sophia Tsoka

Abstract: Cancer is a complex disease with significant social and economic impact. Advancements in high-throughput molecular assays and the reduced cost for performing high-quality multi-omic measurements have fuelled insights through machine learning. Previous studies have shown promise on using multiple omic layers to predict survival and stratify cancer patients. In this paper, we develop and report a Supervised Autoencoder (SAE) model for survival-based multi-omic integration, which improves upon previous work, as well as a Concrete Supervised Autoencoder model (CSAE) which uses feature selection to jointly reconstruct the input features as well as to predict survival. Our results show that our models either outperform or are on par with some of the most commonly used baselines, while either providing a better survival separation (SAE) or being more interpretable (CSAE). Feature selection stability analysis on our models shows a power-law relationship with features commonly associated with survival. The code for this project is available at: https://github.com/phcavelar/coxae .