UCLA

Understanding how and at what scale features emerge from complex natural and social systems has presented daunting challenges for scientists and society. Nevertheless, a unifying approach that spans different systems can offer more comprehensive insights. In this respect, network science provides a common ground and a set of abstractions that has led to important advances in understanding structures, visualizing flows, and establishing predictive models for complex systems. This dissertation studies drug interactions and cardiovascular systems by modeling them as networks to identify and investigate emergent properties. Interaction-based networks play an important role in explaining the characteristics of systems that arise as collective behaviors or responses among different components. Here, we introduce a theoretical framework for the categorization of higher-order interactions built upon the conceptual understanding of net interactions (arising from effects of all levels) and emergent interactions (relative to all lower-order effects). Analyzing and dissecting drug interactions to establish this methodology, we conclude that emergent interactions are common in three-drug combinations, and we observe greater amounts of antagonism with three drugs compared to the pairwise drug combinations. This framework provides promising applications to uncover emergent phenomena into other complex systems with many interacting components such as food webs and social systems. The subsequent parts of this dissertation deal with the vascular system. In particular, we present our recent findings on pervasive asymmetry patterns in cardiovascular branching. Next, we explore the role of optimality principles as well as random branching subject to various spatial constraints from local to intermediate to global in the emergence of these asymmetric patterns. The investigation of spatial scales is important because spatial constraints have been shown to play a key role in the organization of foundationally similar systems such as the distribution of income levels across cities. Overall, this integrative approach focuses on transferring a conceptual framework and knowledge across different fields to help elucidate and predict structures, dynamics, and general behaviors in complex systems.