Presentation at Spring 2014 ACS Meeting in Dallas
Post date: Mar 21, 2014 8:58:28 PM
Conformational and Alchemical Pathways in Extended Ensemble Binding Free Energy Calculations
Monday, March 17, 2014 11:15 AM
Tracing Pathways in Biomolecular Simulation (08:30 AM - 11:45 AM)
Peng He, Lauren Wickstrom, Nanjie Deng, Ronald M Levy
Binding processes involve spatial repositioning of the ligand relative to the receptor. The thermodynamic consequences of such translational and rotational motions, although well understood theoretically, are often misrepresented in applied work. We will briefly review the theory and best practices relating to this aspect of the field. One step up in complexity are variations in the amplitude of motion and redistribution of population among internal degrees of freedom of the binding partners. Such conformational reorganization processes often constitute the main bottleneck for the convergence of binding free energies. Obtaining accurate estimates of reorganization free energies is also very challenging as these are more sensitive to limitations of force fields than interaction energy contributions. In our view, the lack of a faithful representation of reorganization processes is at the basis of the mixed reputation of physical models of binding in pharmaceutical applications. We will review our work in this area and show that replica exchange extended ensemble techniques are particularly suitable for obtaining converged results, and how these, combined with accurate experimental sets, can be deployed on a large scale to validate and improve energy functions. In extended ensemble calculations, the system moves in conformational space as well as alchemical space. The concept of binding pathways in this physical and alchemical joint space is key to the design of efficient binding free energy algorithms and to the understanding of the physical and kinetic mechanisms of binding.
Movies of trajectories shown as part of the presentation:
In this segment the complex undergoes a unbinding-binding-unbinding transition. Note the significant amount of conformational motion. Also the protein and the ligand "freeze-up" when they interact. The concomitant entropic loss plays an important role in regulating binding.
In this experiment the monomers were started facing the "wrong way". One of the monomers slowly turns around and forms native contacts with the other monomer.
In this experiment one of the monomers started as unfolded whereas the other served as a template. During the simulation the first monomer underwent both folding and binding.
In this experiment both monomers were started in a random unfolded conformation. One of the monomers folded into a helix conformation. The second monomer remains unfolded. The simulation is ongoing. Based on the experiment above we expect the dimer to eventually form.