Networks as Biomarkers: Uses and Purposes

Abstract

Networks-based approaches are often used to analyze gene expression data or protein-protein interactions but are not usually applied to study the relationships between different biomarkers. Given the clinical need for more comprehensive and integrative biomarkers that can help to identify personalized therapies, the integration of biomarkers of different natures is an emerging trend in the literature. Network analysis can be used to analyze the relationships between different features of a disease; nodes can be disease-related phenotypes, gene expression, mutational events, protein quantification, imaging-derived features and more. Since different biomarkers can exert causal effects between them, describing such interrelationships can be used to better understand the underlying mechanisms of complex diseases. Networks as biomarkers are not yet commonly used, despite being proven to lead to interesting results. Here, we discuss in which ways they have been used to provide novel insights into disease susceptibility, disease development and severity.

Keywords: biomarkers’ connectivity; integrative biomarker; network analysis; precision medicine.

Figure 1

Networks of biomarkers showcase the underlying relationships between biomarkers of different natures. Data gathered from different sources (such as DNA sequencing, blood samples and electronic medical records) can be studied together as a whole system to shed light onto the molecular mechanism of a disease and help the diagnosis process. Here, biomarkers obtained by the same process are shown with the same colors. Biomarkers of different natures can be positively or negatively correlated.