English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Molecular data from tumor profiles is high dimensional. Tumor profiles can becharacterized by tens of thousands of gene expression features. Due to the sizeof the gene expression feature set machine learning methods are exposed tonoisy variables and complexity. Tumor types present heterogeneity and can besubdivided in tumor subtypes. In many cases tumor data does not include tumorsubtype labeling thus unsupervised learning methods are necessary for tumorsubtype discovery. This work aims to learn meaningful and low dimensionalrepresentations of tumor samples and find tumor subtype clusters while keepingbiological signatures without using tumor labels. The proposed method namedLatent Kernel Feature Selection (LKFS) is an unsupervised approach for geneselection in tumor gene expression profiles. By using Autoencoders a lowdimensional and denoised latent space is learned as a target representation toguide a Multiple Kernel Learning model that selects a subset of genes. By usingthe selected genes a clustering method is used to group samples. In order toevaluate the performance of the proposed unsupervised feature selection methodthe obtained features and clusters are analyzed by clinical significance. Theproposed method has been applied on three tumor datasets which are Brain, Renaland Lung, each one composed by two tumor subtypes. When compared with benchmarkunsupervised feature selection methods the results obtained by the proposedmethod reveal lower redundancy in the selected features and a better clusteringperformance.

Unsupervised Feature Selection for Tumor Profiles using Autoencoders and Kernel Methods