TY - JOUR TI - Inferring models and structure from biological data: networks and pathways DO - https://doi.org/doi:10.7282/t3-rnnm-8y04 PY - 2020 AB - Physiological functions are driven by the emergent behaviors of many individual components, whether they are gene, protein, or metabolic interactions. These interactions form biochemical pathways and interaction networks which then lead to more complex cellular or organismal level behaviors that are not knowable from the characteristics of an individual component of that system. Knowing whether a gene is being expressed or knowing the structure of a protein does not necessarily imply the physiological function of either, but within the context of a meaningful biological system, we can infer more complex behaviors. In the enclosed dissertation we present multiple approaches to contextualize biological components into more complex systems. These methods include utilizing Boolean networks to model interactions in a qualitative manner, as well as analyzing expression data in the context of biochemical pathways. We use two distinct approaches for understanding biological systems: We utilize evolutionary algorithms to understand the origin and development of complex systems. This evolutionary framework enables a better understanding of complex network structures as well as evolutionary strategies used in the development of complex biological systems. We additionally propose a data-driven approach for interrogating gene expression within the context of biochemical pathways. We utilize a novel method for detecting circadian genes and map these genes onto physiologically functional pathways. We utilize this data to validate methods for constructing a Boolean network to infer the causal relationships which exist within gene pathways. This analysis will improve the applications of high throughput data analysis for the purpose of identifying critical components of complex biological systems. KW - Evolution KW - Biomedical Engineering LA - English ER -