Network analysis to understand gene regulation: across spatial scales and toward multi-level models

Current applications of network theory to biology have generally included two types of networks: spatially embedded ones, such as chromatin networks or tissue networks, and abstract representations of interactions, such as protein interactomes or interactions between drugs and their targets. We review recent network theory approaches to epigenomics and argue that expanding the formalisms that we use will be necessary to create more realistic, integrated models that better represent our understanding of complex biological entities and processes, such as gene regulation. On one hand, we need to view intracellular chromatin networks in the context of tissues to improve models of cell interactions. On the other hand, we need to move beyond transcriptional regulation and toward multiomics. We provide several examples of how network formalisms can support this endeavor, especially through applications in cancer research.