Computational Analysis of Bioimaging and Biomolecular Structures to Investigate the Morphology and Conformational Dynamics of Biological Systems Over Multiple Scales

Abstract

Biological data is filled with rich information about complex and multifaceted processes, making it challenging to analyze and extract useful quantitative information. The purpose of this work is to develop and apply computational tools, such as structural modeling or performing molecular dynamics simulations, in order to study biological systems and obtain insights on observed physical events. In the first project, we developed a method to build 3-dimensional beads-on-chain (BOC) models from images of the zebrafish neuroepithelium to measure structural features of the neuroepithelium tissue. We built the BOC models from images of wild type and mutated embryos and designed a method to automatically measure relevant phenotypic characteristics, which identify the physical effect of a gene mutation during a 9-hour span of embryonic development. To further demonstrate the method, we applied it to magnetic resonance images of the human fetal brain. In the second project, we investigated the mechanism of the T cell receptor (TCR) catch bond mechanism that has been observed in single-molecule experiments when sub-20 pN force is applied to the TCR. The catch bond involves an increase in bond lifetime of the TCR and its ligand. We performed all-atom molecular dynamics simulations on several TCR complexes bound to ligands with experimentally known differences in T cell activation. We applied either no, low, or high load on the complexes and developed methods to analyze the trajectories. With this data, we identified structural features that result in the mechanical response of the TCR. In all, this dissertation involves studying the mechanisms of two important biological processes – tissue morphology in neural development and conformational dynamics of the TCR – by applying shared concepts of mechanics and computational analysis. The methods developed for each project provide novel ways to study the respective systems and the results we found will aid in future studies that utilize computational techniques to analyze biological systems.