Computational studies for prediction of protein folding and ligand binding

• Main Author: Luo, Lingqi
• Language: en_US
• Published: 2018
• Subjects: Bioinformatics
• Online Access: https://hdl.handle.net/2144/27366

This dissertation comprises four projects. I) Glycosylation is a post-translational modification that affects many physiological processes, including protein folding, cell interaction and host immune response. PglC, a phosphoglycosyl transferase (PGT) involved in the biosynthesis of N-linked glycoproteins in Campylobacter jejuni, is representative of one of the structurally simplest members of the small bacterial PGT family. The research utilizes sequence similarity network and evolutionary covariance studies to identify the catalytic core of PglC, followed by modeling its three-dimensional structure using the covariance as constraints. II) Rapid growth of fragment-based drug discovery necessitates accurate fragment library screening for targets of interest, finding strong binders with specific binding. While many high-resolution biophysical methods for fragment screening work well, docking-based virtual screening is highly desired due to the speed and cost efficiency. Beyond the key performance-determining factors like score function and search method, the goal is to learn from the experimental fragment bound structures in the PDBbinder database and to evaluate the profile of side-chain flexibility in the interface and its contribution to docking performance. III) Protein docking procedures carry out the task of predicting the structure of a protein–protein complex starting from the known structures of the individual protein components. However, the structure of one or both components frequently must be obtained by homology modeling based on known structures. This work presents a benchmark dataset of experimentally determined target complexes with a large set of sufficiently diverse template complexes identified for each target. The dataset allows developers to test their algorithms combining homology modeling and docking, in order to determine the factors that critically influence the prediction performance. IV) Human Eukaryotic Initiation Factor 4AI (heIF4AI) is the enzymatic component of a highly efficient complex, heIF4F. Its helicase activity binds and unwinds the secondary structure of mRNA at the 5' end and thus plays a crucial role in translation initiation. This research focuses on the C-terminal domain of heIF4AI and investigates its potential as an anti-cancer target by integrating the approaches of solvent mapping, docking, crystallization and NMR.