Bayesian multi-domain learning for cancer subtype discovery from next-generation sequencing count data

TL;DR

This paper introduces a Bayesian multi-domain learning model that effectively integrates diverse NGS count data for accurate cancer subtype discovery, addressing overdispersion and limited sample sizes in complex diseases.

Contribution

The paper presents a novel hierarchical negative binomial factorization model for multi-domain NGS data, improving cancer subtyping accuracy with small sample sizes and reducing negative transfer.

Findings

Effective multi-domain learning demonstrated on TCGA datasets

Outperforms existing transfer learning methods in cancer subtyping

Handles overdispersed count data accurately

Abstract

Precision medicine aims for personalized prognosis and therapeutics by utilizing recent genome-scale high-throughput profiling techniques, including next-generation sequencing (NGS). However, translating NGS data faces several challenges. First, NGS count data are often overdispersed, requiring appropriate modeling. Second, compared to the number of involved molecules and system complexity, the number of available samples for studying complex disease, such as cancer, is often limited, especially considering disease heterogeneity. The key question is whether we may integrate available data from all different sources or domains to achieve reproducible disease prognosis based on NGS count data. In this paper, we develop a Bayesian Multi-Domain Learning (BMDL) model that derives domain-dependent latent representations of overdispersed count data based on hierarchical negative binomial factorization for accurate cancer subtyping even if the number of samples for a specific cancer type is small. Experimental results from both our simulated and NGS datasets from The Cancer Genome Atlas (TCGA) demonstrate the promising potential of BMDL for effective multi-domain learning without "negative transfer" effects often seen in existing multi-task learning and transfer learning methods.