Thesis defence

Random Forest for short biomarker signature discovery in cancer.



Mr Ahmed Debit will publicly defend his thesis entitled "Random Forest for short biomarker signature discovery in cancer".

 

Summary

Biomarker signatures in cancer are generally defined as a single or a combined alteration of genes associated with a defined biological tumor phenomenon. These signatures are providing clinicians with significant information to improve our understanding of cancer biology. Furthermore, with a validated specificity and sensitivity, molecular signatures can be used as a clinical tool for screening, predicting progression, and treatment response. Next generation sequencing technologies allow us to measure the expression profile of the genes with high resolution. Thanks to advances in bioinformatics techniques and the emergence of advanced statistical approaches, deriving biomarker signatures by mining such data has become increasingly popular.

In this thesis, we propose a comprehensive pipeline for Short Biomarker Signature Discovery (sBSD) in cancer. The major challenge is to design a stable set of genes, as small as possible, that accurately predicts learned patterns. Furthermore, this thesis covers important aspects related to the application of machine learning techniques and bioinformatics approaches to gene expression data. Our proposed strategy focuses on the stability of methods, clinical interpretation of results, and applicability in the context of gene expression data.

 

Practical information

Defence will take place on September 27th 2020 at 13:00, Auditoire L. Frédéricq, Bâtiment B34, au Sart Tilman (access and map) or to all via lifesize : https://call.lifesizecloud.com/5397980 (Code : 271020#).

Share this news