BIOCHEMICAL PROFILE-BASED COMPUTATIONAL INFERENCE OF PROTEIN COMPLEXES
HU Zhong-ming · TSpace (University of Toronto) · 2020
Protein complexes are key macromolecular machines of the cell, but their description remains incomplete. Our group and others previously reported an experimental strategy for global characterization of native protein assemblies based on chromatographic fractionation of biological extracts coupled to precision mass spectrometry analysis (chromatographic fractionation–mass spectrometry, CF–MS), but the resulting data are challenging to process and interpret. In this thesis, I describe EPIC (elution profile-based inference of complexes), a software toolkit for automated scoring of large-scale CF–MS data to define high-confidence multi-component macromolecules from diverse biological specimens. The software toolkit EPIC is “plug-and-play”, connects to public repositories for automatic data processing, and can be adopted productively to explore the network biology of any model system with little computational expertise required. The optimized CF-MS pipeline and EPIC data analysis workflows described in this thesis can be used to study different biological specimens, including diverse model organisms, to chart protein complexes on a global scale to expand our knowledge of macromolecular networks and their association with physiology, development, evolution and disease. Beyond providing a powerful framework to interpret CF-MS data, as a case study, I used EPIC to map the global interactome of Caenorhabditis elegans (WormMap), an important genetic model, comprising 612 putative multi-protein complexes linked to diverse biological processes. These encompassed new subunits for previously annotated protein complexes as well as novel assemblies seemingly unique to nematodes that we verified using stringent benchmarking criteria as well as by independent orthogonal affinity-purification mass spectrometry validation experiments. To my knowledge, this is the first biochemically-based large-scale map of nematode protein complexes, which provides a rich platform for hypothesis-driven mechanistic investigations of animal biology. The major two outcomes of this dissertation consist of a tool (EPIC) and a knowledgebase (WormMap), which should serve as lasting resources for the broader biological research community.