Our research is focused on the development of novel computational methods to solve biological problems in the field of medicine. We believe that computational work must be driven to match experimental data. We use statistical mechanical methods to understand the thermodynamics of binding, with specific application to molecular design tools for developing effective new therapeutics.
On the theoretical side, we are interested in computing entropy and mutual information from high-order correlations. We also devise methods to probe the surfaces and binding sites of proteins to find druggable targets, by identifying binding hot spots. This work is applied through collaborations with a number of drug discovery projects at the University of Cambridge and elsewhere, applying rational methods to design small-molecule inhibitors for diseases such as cancer, malaria and tuberculosis
The concepts of entropy and mutual information (MI) are fundamental aspects of statistical mechanics. However, whilst entropy is a key driver in chemical and biological processes, it remains difficult to model and is consequently ill understood. Various entropy estimation techniques have been developed, with the most direct being based on information theoretic approaches.
Computational molecular design is a useful tool in modern drug discovery. Virtual screening is an approach that evaluates individual members of compound libraries. In contrast, design approaches construct compounds by combining scaffolds and sidegroups to optimise the calculated binding affinity.
Inhomogeneous fluid solvation theory (IFST) is a statistical mechanical framework for calculating the effect of a solute on the free energy of the surrounding solvent relative to its bulk state. The solute can be a protein, peptide, or small molecule and the solvent is commonly water. One of the useful features of IFST is that the free energy changes are calculated for small subvolumes surrounding the solute and this allows the contribution of different regions of space to be calculated and visualized.
The druggability of a protein target is defined as the relative ease or difficulty of developing a small molecule that can effectively modulate the protein’s activity in vivo. The ligandability of a protein is defined as the relative ease or difficulty of developing a small molecule that can inhibit the protein in vitro. This is an important difference, as there are complex pharmacodynamic and pharmacokinetic factors that influence druggability but not ligandability.
Water is one of the simplest molecules in existence, but also one of the most important in biological and engineered systems. However, understanding the structure and dynamics of liquid water remains a major scientific challenge. Molecular dynamics simulations of liquid water can be used to calculate the radial distribution functions (RDFs), the relative angular distributions, and the excess enthalpies,entropies, and free energies.
Shrewd selection of screening compounds is one of the most vital enabling steps in the drug discovery process. A screening library must have the correct balance of properties such as molecular weight and lipophilicity. An ideal hit molecule must also be amenable to chemical elaboration, show reasonable levels of cell permeability, and have a range of commercially available analogues, some of which have also been tested in the same assay.