Structure Based Drug Design

The three-dimensional structure of a target protein, especially with bound ligand, is a valuable source of information to guide drug discovery. A number of structure-based drug design tools, complemented by ligand-based approaches, can be used for the design and Optimization of new compounds.

Structure Based Drug Design

• Our structural biology team allows us to generate protein constructs and obtain X-ray crystallography data on behalf of customers for targets lacking structural information and to elucidate the binding mode of representative leads to drive structure-based optimisation.
• We have extensive experience in generating homology models based on related proteins for any targets that lack structural data and prove difficult to crystallize.
• Our expert computational team analyses all available structural information and shares results and 3D visualization tools with the medicinal chemistry teams to encourage idea generation and improve compound design.
• Flexible ligand docking and 4D docking, to account for protein flexibility, are applied to prioritize design ideas and a 3D fully-interactive ligand editor can be used for ligand optimization in the binding site.
• A range of modern techniques are used to assess and rank the affinity and stability of binding poses, including short trajectory MD simulations, MM/PBSA and Grid Free Energy calculations.
• We can apply water assessment technology to identify favourable and unfavourable regions of solvent thermodynamics in the binding site. Introduction of polar functionalities to a lead compound, based on this understanding of the water structure, can often benefit physical properties as well as potency and specificity.

We are passionate about acquiring and using cutting-edge tools to advance drug discovery and are always looking for ways to improve our capabilities. We have therfore initiated a collaboration with a leading academic expert in Free Energy Perturbation (FEP) calculations. As part of this ongoing effort, we are investigating the factors affecting the accuracy of protein-ligand binding free energy estimates. In a recent study, we highlighted the impact of the initial crystal structure on the resulting free energy predictions and emphasized the importance of rare event sampling and long-timescale dynamics when running FEPs. We are currently working on finding computational solutions to improve the prediction accuracy.