Date of Award

12-2024

Degree Type

Thesis

Degree Name

Master of Science (MS)

Department

IT Innovation

First Advisor

Dr. Dhundy (Kiran) Bastola

Abstract

Classification of patients with cancer is a crucial first step for clinicians to understand disease characteristics and manage effective treatments. C3MOD is a Python-based tool developed to advance cancer research by leveraging multiomics data for cancer subtype identification and characterization. The tool employs unsupervised machine learning algorithms, including KMeans, Similarity Network Fusion (SNF), and hierarchical clustering, to stratify patients based on datasets including miRNA, mRNA, and DNA methylation profiles. Its customizable framework allows researchers to analyze various cancer types, providing insights into survival outcomes, mutation burden, and immune characterization of identified subtypes.

C3MOD addresses the challenges in multiomics analysis, such as data integration and flexibility, by offering a user-friendly and scalable solution. It provides an efficient and accurate means of identifying cancer subtypes, which can facilitate precision medicine strategies tailored to individual patient profiles.

The tool has been validated using TCGA datasets for 30 cancer types, demonstrating its robustness and utility in identifying meaningful subtypes and biomarkers, such as GNG4 in lung 2 adenocarcinoma (TCGA-LUAD). This biomarker has been linked to poor survival outcomes, further underscoring the clinical relevance of C3MOD.

Comments

The author holds the copyright to this work, any reuse or permissions must be obtained from them directly.

Download

Share

COinS