SCHOOL OF COMPUTER SCIENCE

The School of Computer Science is pleased to present…

 

Date: Friday April 17, 2020

Time:  1:00pm – 4:00pm

ZOOM Meeting URL: https://zoom.us/j/866457095?pwd=YjFodVFnc2s1bmczZEhac1pEWmFmZz09

Meeting ID: 866 457 095

Password: Request password at csgradinfo@uwindsor.ca

 

Identifying biomarkers that can help predict a cancer class with high accuracy is at the heart of breast cancer studies. Genomics data have been shown to carry useful information for breast cancer diagnosis and prognosis, as well as uncovering the disease’s mechanism. Machine learning methods are powerful tools to find such information. In this work, we aim at identifying meaningful biomarkers for predicting the survivability of breast cancer patients receiving a certain treatment.

We have applied and developed feature selection methods on gene-expression data to derived gene-signatures, where the initial genes are collected concerning the mechanism of some drugs used breast cancer therapies. We have also proposed two network-based approaches to identify sub-network biomarkers. They integrate gene-expression data with protein-protein interactions during the optimal subnet searching process and use cancer-related genes and pathways to prioritize the extracted sub-networks. Several heuristics have been implemented to deal with the huge search space and class-imbalanced data. We have found potential gene signatures and sub-network biomarkers that are biologically meaningful and can yield significantly high accuracy in prediction. Among the potential biomarker extracted there are well-known breast-cancer-related genes such as TP53, MYC, PIKCA3, SPY1, ERBB4, STAT3, STAT5A, FGFR2.

 

Thesis Committee: 

 PhD Dissertation Defense Announcement