Overview

Cancer is the second leading cause of death in the United States and because of the complexity and heterogeneity of the disease there are no current cures available. Federal funding for cancer research over the past decade has emphasized the investigation of biomarkers for specific cancer types. Clinical biomarkers (such as family history, predisposition, lifestyle, environmental exposure, etc) and molecular biomarkers (such as genetic mutations, genome-wide methylation, gene expression, etc) utilized together provide patient-specific insights to predict, diagnose and treat cancer. Genomic biomarkers are particularly important because cancer originates due to several acquired genetic mutations. Sequencing of cancer genomes has been successful (as reported in scientific literature) in characterizing the molecular landscape of cancerous cells and discovering biomarkers. Thus, the purpose of our information visualization project is to visualize massive clinical and genomic datasets of cancer—kidney cancer in particular—to facilitate the discovery of robust biomarkers.

The project had the following aims:

Aim 1: Display relationships between multiple heterogeneous genomic datasets of kidney cancer

Aim 2: Allow for exploratory insights into the possible genomic causes/predictors (gene expression, and DNA methylation) of kidney cancer

Aim 3: Display clinical and genomic markers for each subtype of kidney cancer

View the visualization here.

Objective

Apply information visualization principles to mutlimodal datasets

Course

CS 4460 (Georgia Tech)

What I Did

D3 programming; Responsive web design

Responsive design should lower the cognitive strain on information processing.

Proper information visualizations can transform data to actionable knowledge to clinical decisions.