Automated Cancer Subtyping via Vector Quantization Mutual Information Maximization

TL;DR

This paper introduces an unsupervised clustering method that uses vector quantization and mutual information maximization to identify cancer subtypes from high-dimensional genetic expression data, refining labels and correlating with survival rates.

Contribution

It presents a novel, label-agnostic clustering approach that adaptively determines the number of cancer subtypes using mutual information maximization on genetic profiles.

Findings

Refines existing cancer subtype labels

High correlation with cancer survival rates

Automatically determines the number of subtypes

Abstract

Cancer subtyping is crucial for understanding the nature of tumors and providing suitable therapy. However, existing labelling methods are medically controversial, and have driven the process of subtyping away from teaching signals. Moreover, cancer genetic expression profiles are high-dimensional, scarce, and have complicated dependence, thereby posing a serious challenge to existing subtyping models for outputting sensible clustering. In this study, we propose a novel clustering method for exploiting genetic expression profiles and distinguishing subtypes in an unsupervised manner. The proposed method adaptively learns categorical correspondence from latent representations of expression profiles to the subtypes output by the model. By maximizing the problem -- agnostic mutual information between input expression profiles and output subtypes, our method can automatically decide a suitable number of subtypes. Through experiments, we demonstrate that our proposed method can refine existing controversial labels, and, by further medical analysis, this refinement is proven to have a high correlation with cancer survival rates.