Harnessing the Power of Integrative Bioinformatics and Machine Learning for Unveiling Biomarkers and Classifying Colorectal Cancer Patients

Candidate

• Mark Fertel, Louisiana State University Health Sciences Center New OrleansFollow
• Chindo Hicks, Louisiana State University Health Sciences Center New OrleansFollow
• Duaa Alawad, Louisiana State University Health Sciences Center New OrleansFollow

Examination Date

Spring 3-18-2025

Degree

Thesis

Degree Program

Bioinformatics

Examination Committee

Dr. Chindo Hicks, Dr. Diptasri Mandal, Dr. Jovanny Zabaleta

Abstract

Colorectal cancer (CRC) remains a major cause of cancer-related

mortality, emphasizing the need for novel diagnostic and prognostic biomarkers.

In this study, we developed an integrative bioinformatics framework combining

RNA-Seq and somatic mutation data from The Cancer Genome Atlas (TCGA)

to identify molecular drivers of CRC. We found 12,899 differentially expressed

somatic mutated genes distinguishing tumors from controls, including CDH3,

ETV4, and COL11A1. Enrichment analysis highlighted key pathways such as

Wnt, p53, and cell cycle signaling. A survival-based comparison identified 2,642

genes linked to chromatin assembly and nucleosome organization. Machine

learning (ML) models were developed for patient classification and survival

prediction, with XGBoost outperforming others. This integrative approach

demonstrates the utility of combining transcriptomic and mutational data with

ML for biomarker discovery, patient stratification, and outcome prediction,

offering valuable insights for CRC clinical management.

Recommended Citation

Fertel, Mark; Hicks, Chindo; and Alawad, Duaa, "Harnessing the Power of Integrative Bioinformatics and Machine Learning for Unveiling Biomarkers and Classifying Colorectal Cancer Patients" (2025). School of Graduate Studies. 10.
https://digitalscholar.lsuhsc.edu/etd_sgs/10