Figure 2.

Potential Approaches to Reclassify Complex Diseases in Network Medicine. Although the optimal approach to integrate multiple types of biological and clinical information into a new classification for disease remains speculative, we have outlined a potentially useful framework. After an appropriate study population is obtained, collection of biological samples and comprehensive phenotyping will be required. Multiple –omics assessments will be performed, including whole genome sequencing, genome-wide gene expression and proteomic analysis, and metabolomic assessment. These data types will be integrated using network-based approaches to identify determinants of the complex disease. In parallel, comprehensive phenotyping will be performed; machine learning (and other) approaches may be used to define disease subtypes. In an iterative process, the disease determinants and disease subtypes will be refined to create a pathophysiology-based disease classification.