Authors: Jasper, Julia B.
Title: Local thermodynamic characterisation of binding sites and protein-ligand interactions
Language (ISO): en
Abstract: With the advent of high-performance computing and progress in fields like structural biology, structure-based drug design (SBDD) has become an essential part in drug discovery projects. Yet, the automated de novo design of the “perfect” ligand to a protein structure remains unreached, even with today’s (re)emergence of powerful machine learning techniques. One of the reasons for this is the difficulty to consider all thermodynamically relevant details of protein-ligand complex formation in the modelling and design process. In classical SBDD approaches, the direct intermolecular interactions between host and guest are usually well accounted for. More subtle aspects include the change in both ligand and protein conformations upon binding and, above all, solvent effects. Investigating these solvent effects, and especially deriving rules how to exploit knowledge about the local thermodynamic properties of protein hydration sites for drug design purposes, was the main objective of this work. To tackle this challenge, the 3D reference interaction site model (3D RISM) was employed to predict the solvent distribution within binding sites as well as local thermodynamic properties of specific hydration sites. The first part of this work hence focused on a large-scale analysis of the thermodynamic signatures of protein hydration sites and their correlation with ligand features. Water replacement rules for use in ligand design and optimisation were derived and exemplified on matched molecular pairs (MMPs). The second part of this work expanded the concept of 3D RISM-derived thermodynamic binding site characterisation to virtual probe sites mimicking specific functional groups whose distribution within a binding site can be calculated by a 3D RISM solute-solute approach. An advanced framework combining RISM-based binding site characterisation with library preparation, docking, and scoring was established that allows to “convert” the probe densities to a selection of promising fragments or small molecules. In a third part, the afore-mentioned concepts were applied together onto three case studies from the challenging field of protein-protein interactions (PPIs) to illustrate the practicability of the developed approaches.
Subject Headings: Wirkstoffforschung
Subject Headings (RSWK): Wirkstoffforschung
Issue Date: 2022
Appears in Collections:Physikalische Chemie

Files in This Item:
File Description SizeFormat 
Dissertation.pdf31.99 MBAdobe PDFView/Open

This item is protected by original copyright

Items in Eldorado are protected by copyright, with all rights reserved, unless otherwise indicated.