Automated docking is an important tool for gaining a mechanistic understanding of the molecular interactions underpinning most biological processes. It supports biomedical applications ranging from drug and chemical probes design to investigating chemical pathways and identifying therapeutic targets for diseases such as cancer and metabolic disorders. Two major challenges for automated docking remain the conformational changes occurring in macromolecules upon ligand binding and the docking highly flexible ligands such as peptides, an area that has gained a lot of interest recently. Most docking programs use rigid models of macromolecules because of the exponential increase of computational complexity and the decreased docking accuracy associated receptor flexibility. Under the previous award, we made significant progress towards docking with flexible receptors. We developed and released a software program allowing the successful cross- docking of flexible ligands into apo conformations of receptors with up to 14 flexible side-chains, and demonstrated that down-weighting the contribution of the receptor internal energy to the scoring function increased docking success rates. For the next award, we propose to: 1) extend the set of motion objects for describing ligand conformational changes beyond rotatable bonds, and motion objects for representing local and global receptor backbone motions such as flexible loops and domain motions; 2) enhance our current search technique and develop a new representation and associate search technique for docking peptides; 3) calibrate and optimize the scoring function sused during docking specifically in the context of docking with receptor flexibility; and 4) make the software useful and usable through clearly written documentation and intuitive graphical user interfaces. Datasets compiled under this project will also be made freely available to the community and provide a benchmark for evaluating docking methods. This Open-Source software development project will be based on best practices in software engineering and result in a modular, component-based software environment with pluggable search techniques and scoring functions, and an extensible set of molecular flexibility descriptors. This new docking engine will be part of the widely used and popular software suite AutoDock. It will be used on the World Community Grid for large virtual screens on therapeutically important targets. The ability to include receptor flexibility in docking simulations will extend the range of biological problems for which automated docking can be successfully applied. Hence, it will impact the research of many medicinal chemists and biologist, and extend the use of computational tools to a wider community of scientists thereby supporting the advancement of biomedical research.
Automated docking is a workhorse for a variety of applications including: rational drug design, lead optimization and the design of chemical probes, however, the rigid representation of macromolecules still used by most docking programs severely limits its range of applications. We made substantial progress toward supporting receptor flexibility during the prior funding period, by developing a practical and usable approach for docking with up to 14 flexible receptor side chains, and we now propose extend the repertoire of descriptors for simulating backbone flexibility in receptors, developing a new protein-peptide docking engine, and optimize scoring functions to increase docking success rates. This new docking programs usable for routine dockings of peptides and docking with receptor flexibility will impact the research of many computational and medicinal chemists and biologist and contribute to our understanding of biological processes and thus significantly impact research in biomedicine.
| Fernández, José A; Xu, Xiao; Sinha, Ranjeet K et al. (2017) Activated protein C light chain provides an extended binding surface for its anticoagulant cofactor, protein S. Blood Adv 1:1423-1426 |
| Ravindranath, Pradeep Anand; Sanner, Michel F (2016) AutoSite: an automated approach for pseudo-ligands prediction-from ligand-binding sites identification to predicting key ligand atoms. Bioinformatics 32:3142-3149 |
| Forli, Stefano; Huey, Ruth; Pique, Michael E et al. (2016) Computational protein-ligand docking and virtual drug screening with the AutoDock suite. Nat Protoc 11:905-19 |
| Ravindranath, Pradeep Anand; Forli, Stefano; Goodsell, David S et al. (2015) AutoDockFR: Advances in Protein-Ligand Docking with Explicitly Specified Binding Site Flexibility. PLoS Comput Biol 11:e1004586 |
