Development and Application of Computational Biology tools for Biomedicine

Biomolecular simulation can be considered as a virtual microscope for molecular biology, allowing to gain insights into the sub-cellular mechanisms of biological relevance at spatial and temporal scales that are difficult to observe experimentally. It provides a powerful tool to link the laws of physics with the complex behavior of biological systems. Dramatic recent advancements in achievable simulation speed and the underlying physical models will increasingly lead to molecular views of large systems. These improvements may largely affect biological sciences. In this thesis, I have applied computational molecular biology approaches to different biological systems using state of the art structural bioinformatics and computational biophysics tools (Chapter 3). My principal focus was on the computational design of molecular imprinted polymers (MIPs), which have recently attracted significant attention as cost effective substitutes for natural antibodies and receptors in chromatography, sensors and assays. I have used molecular modelling in the optimization of polymer compositions to make high affinity synthetic receptors based on Molecular Imprinting. In particular, I developed a new free of charge protocol that can be performed within just few hours that outputs a list of candidate monomers which are capable of strong binding interactions with the template. Furthermore, I have produced a new computational method for the calculation of the ideal monomer: template stoichiometric ratio to be used in the lab for the MIPs synthesis. These protocols have been implemented as a webserver that is available at http://mirate.di.univr.it/. In parallel, I have also investigated the modelling of much more complex MIPs systems by the introduction of some factors e.g. solvent and cross-linker molecules that are also essential in the polymerisation process. A novel algorithm, which mimics a radical polymerization mechanism, has been written for application in the rational design of MIPs (Chapter 4). Moreover, I have been involved in the field of computational molecular biomedicine. Indeed, in Chapters 5 and 6 I describe the work done in collaboration with two labs at the Department of Neuroscience, Biomedicine and Movement Sciences, University of Verona. In Chapter 5, starting from unpublished experimental data I have computationally characterized the interaction of ACOT8 with HIV-1 Nef accessory protein. I have performed a detailed structural and functional characterization of these two proteins in order to infer any possible functional details about their interactions. The bioinformatics predictions were then confirmed by wet-lab experiments. I have also carried out a detailed structural and functional characterization of two pathogenic mutations of AGT-Mi (Chapter 6). In particular, I have used classical molecular dynamics (MD) simulations to study the possible interference with the dimerization process of AGT-Mi exerted by I244T-Mi and F152I-Mi mutants. Those variants are associated with Primary Hyperoxaluria type 1 disease. In Chapter 7, I present the coarse-grained MD simulations of Membrane/Human ileal bile-acid-binding protein Interactions. This study was carried out in collaboration with the NMR group at the University of Verona and it is a part of an extensive research aimed at better understanding of the main biomolecular interactions in crowded cellular environments. MD simulations results were in agreement with experimental findings.