A “Complex” Problem: Leveraging spatial and temporal information to investigate dynamic protein-protein interactions

Abstract

Large omic datasets with dynamic characterizations of biomolecules across space and time hold immense potential to provide valuable insights into biological processes. However, parsing this information poses significant challenges in the interpretation of multidimensional data. The field of protein-protein interactions would especially benefit from advancements in visualizing, analyzing, and integrating spatially and temporally dynamic information. To address this need, I developed VISTA (Visualizing Interactions in Space and Time Analysis tool), a computational program to explore dynamic protein interaction networks through animated visuals and network reports, with automatic integration of published knowledge on protein localizations, functions, and interactions. The program is freely available for download at https://github.com/samvidav/VISTA. To demonstrate the utility of VISTA, I performed immunoprecipitation and mass spectrometry analyses of three viral proteins of the spatially and temporally regulated pathogen human cytomegalovirus (HCMV) and used the integrated analysis capabilities of VISTA to better define their biological roles. Early in infection, pUL13 interacts with actin remodeling proteins, potentially contributing to cellular rounding, while at later timepoints, pUL37 and pUL13 likely have cooperative roles in the mitochondria to disrupt mitochondrial morphology through the MICOS complex. pUL13 additionally interacts with proteins in the unfolded protein response pathway in the virus assembly complex. Finally, a preliminary investigation also identified interactors of pUL82 involved in viral gene expression and cell cycle control. In Chapter 2, I expanded upon VISTA bioinformatic analysis of proteomics data to infer protein complex conservation and divergence in the heart proteomes of four cardiac model organisms, uncovering species-specific immune and metabolic protein complexes, as well as conserved roles for protein expression and localization in all species studied. Overall, this thesis synthesizes a range of computational and experimental methods for the study of protein interactions in diverse biological contexts and extends these tools to other researchers for broad application in the field.