Using Data Science to Understand Complexity and Quantify Heterogeneity in the Onset and Progression of Chronic...

Novel discoveries in genomics and other omics sciences are rapidly redefining our understanding of health and disease as well as advancing the development of targeted therapeutics for improving health outcomes. The scale of these findings, fueled by vast increases in computing power and new techniques in data analytics, easily supersedes that of phenomena observed using more traditional research approaches. Until recently, the classification and diagnosis of disease has involved rather subjective processes, whereby signs and late symptom patterns are linked with clinical outcomes. However, symptom patterns, disease trajectories, and health outcomes are complex entities characterized by a wide range of clinical manifestations and progression patterns. The burgeoning fields of data science and bioinformatics are opening opportunities for nurse scientists to quantify disease heterogeneity by defining and categorizing disease phenotypes and endotypes. Nurse scientists and clinicians can play a critical role in engaging patients and the larger scientific community in these efforts. The purpose of this article is to provide an introduction to concepts critical to understanding and quantifying heterogeneity in the onset and progression of chronic disease. To present and exemplify key concepts, we (1) discuss evidence for heterogeneity in the onset and progression of Type 1 diabetes, (2) link emerging research approaches in data science with principles in network science and systems biology to lay the groundwork for stratifying subclinical and advanced chronic disease, thus expanding the purview of symptom science, and (3) describe the computational skills needed to engage in these analyses.