Date of Award

1-1-2018

Document Type

Dissertation

Degree Name

Ph.D.

Organizational Unit

Physics and Astronomy

First Advisor

Kingshuk Ghosh, Ph.D.

Second Advisor

Markus Schneider, Ph.D.

Third Advisor

Todd Blankenship

Fourth Advisor

Michelle Knowles

Fifth Advisor

Mark Siemens

Keywords

Biophysics, MaxCal, Modeling, Morphology, Protein, Stochasticity

Abstract

Stochasticity, disorder, and noise play crucial roles in the functioning of many biological systems over many different length scales. On the molecular scale, most proteins are envisioned as pristinely folded structures, but intrinsically disordered proteins (IDPs) have no such folded state and still serve distinct purposes within the cell. At the scale of gene regulation, realistic in vivo conditions produce stochastic fluctuations in gene expression that can lead to advantageous bet-hedging strategies, but can be difficult to characterize using a deterministic framework. Even at the organismal scale, germband extension (GBE) in Drosophila melanogaster embryos systematically elongates the epithelial tissue using cell intercalation, but leaves cells in highly heterogenous geometries. Throughout this work, we will demonstrate that these characteristics are not just arbitrary artifacts to be glossed over, but are actually very intentional frameworks that are harnessed by the respective systems to their own advantage.

In some cases, they can also be harnessed by researchers to better characterize or even control the system through various biophysical techniques. In the case of IDPs, we will introduce an analytical model that can predict the conformational size of these disordered proteins and identify specific "hot spots" in their sequences that hold significant influence over the shape (and therefore function) of each protein. To better understand stochastic gene expression, the power of stochastic inference methods will be put on display, specifically methods using modeling systems based on the principle of Maximum Caliber. These require no direct knowledge about the architecture of the underlying genetic network and make quantitative predictions using the entire content of experimentally realistic time-series data. Finally, we will break down the process of GBE using node-based Monte Carlo simulations to show that while anisotropic tension is enough to qualitatively reproduce convergent extension, competing active extension mechanisms must be introduced in each cell to achieve heterogeneous cell configurations and quantitative agreement with experiment. These studies will collectively demonstrate that randomness and fluctuation do not always imply disarray and intractibility, but instead can convey adaptability and possibility. As such, these characteristics should be embraced by the field of biophysics.

Publication Statement

Copyright is held by the author. User is responsible for all copyright compliance.

Rights Holder

Taylor Emil Firman

Provenance

Received from ProQuest

File Format

application/pdf

Language

en

File Size

194 p.

Discipline

Biophysics



Share

COinS