Unsupervised Feature Selection for Tumor Profiles using Autoencoders and Kernel Methods

TL;DR

This paper introduces LKFS, an unsupervised method combining autoencoders and kernel learning to select relevant genes and discover tumor subtypes from high-dimensional gene expression data without labels.

Contribution

The paper presents LKFS, a novel unsupervised gene selection technique that improves clustering of tumor subtypes by preserving biological signatures and reducing feature redundancy.

Findings

Lower redundancy in selected features

Improved clustering performance

Effective across multiple tumor datasets

Abstract

Molecular data from tumor profiles is high dimensional. Tumor profiles can be characterized by tens of thousands of gene expression features. Due to the size of the gene expression feature set machine learning methods are exposed to noisy variables and complexity. Tumor types present heterogeneity and can be subdivided in tumor subtypes. In many cases tumor data does not include tumor subtype labeling thus unsupervised learning methods are necessary for tumor subtype discovery. This work aims to learn meaningful and low dimensional representations of tumor samples and find tumor subtype clusters while keeping biological signatures without using tumor labels. The proposed method named Latent Kernel Feature Selection (LKFS) is an unsupervised approach for gene selection in tumor gene expression profiles. By using Autoencoders a low dimensional and denoised latent space is learned as a target representation to guide a Multiple Kernel Learning model that selects a subset of genes. By using the selected genes a clustering method is used to group samples. In order to evaluate the performance of the proposed unsupervised feature selection method the obtained features and clusters are analyzed by clinical significance. The proposed method has been applied on three tumor datasets which are Brain, Renal and Lung, each one composed by two tumor subtypes. When compared with benchmark unsupervised feature selection methods the results obtained by the proposed method reveal lower redundancy in the selected features and a better clustering performance.