<p> Modern biological research is currently canalized into two main modes of research: detailed, mechanistic descriptions, or big data collection and statistical descriptions. The former has the advantage of being conceptually tractable and fitting into an existing scientific paradigm. However, these detailed descriptions can suffer from an inability to be understood in the larger context of biological phenomena. On the other hand, the big data approaches, while closer to being able to capture the full depth of biological complexity, are limited in their ability to impart conceptual understanding to researchers. We put forward examples of an intermediate approach. The goal of this approach is to develop models which can be understood as abstractions of biological phenomena, while simultaneously being conducive to modeling and computational approaches. Firstly, we attempt to examine the phenomenon of modularity. Modularity is an ubiquitous phenomenon in biological systems, but its etiology is poorly understood. It has been previously shown that organisms that evolved in environments with lower levels of stability tend to display more modular organization of their gene regulatory networks, although theoretical predictions have failed to account for this. We put forward a neutral evolutionary model, where we posit the process of genome expansion through gene duplications acts as a driver for the evolution of modularity. This process occurs through the duplication of regulatory elements alongside the duplication of a gene, causing sub-networks to be generated which are more tightly coupled internally than externally, which gives rise to a modular architecture. Finally, we also generate an experimental system by which we can verify our model of the evolution of modularity. Using a long term experimental evolution setup, we evolve <i>E. coli</i> under fluctuating temperature environments for 600 generations in order to test if there is a measurable increase in the modularity of the gene regulatory networks of the organisms. This data will also be used in the future to test other hypotheses related to evolution under fluctuating environments. The second such model is a computational model of the properties of bacterial growth as a function of temperature. We describe a model composed of a chain of enzyme like actions, where the output of each enzyme in the chain becomes the substrate of the following enzyme. Using well known temperature dependence curves for enzyme activity and no further assumptions, we are then able to replicate the salient properties of bacterial growth curves at varying temperatures, including lag time, carrying capacity, and growth rate. Lastly, we extend these models to attempt to describe the ability of cancer cells to alter their phenotypes in ways that would be impossible for normal cells. We term this model the <i>phenotypically pliant cells</i> model and show that it can encapsulate important aspects of cancer cell behavior.</p><p>
Identifer | oai:union.ndltd.org:PROQUEST/oai:pqdtoai.proquest.com:10798623 |
Date | 02 March 2018 |
Creators | Biro, Daniel |
Publisher | Yeshiva University |
Source Sets | ProQuest.com |
Language | English |
Detected Language | English |
Type | thesis |
Page generated in 0.0017 seconds