I am just curious if is possible to relate or derive the Koff (or eventually Kd) of ligand-receptor complex on the basis of the residence time observed in the binding site during Molecular dynamics simulations.
I have a series of weak binders and I would like to do some MD simulations in order to evaluate the residence time in the binding site during MD. So, I wonder if I could use some unbiased or enhanced MD methods to derive Koff or Kd from the residence time of each compound.
1. Calculate the protein-ligand binding free energy using free energy perturbation or linear response approximation (or its simplified version: linear interaction energy); you will need to run two sets of MD simulations: protein-ligand complex in solution and free ligand in solution.
2. Calculate Kd from the protein-ligand binding free energy (deltaG = RTlnKd).
Hi Martin and thanks for your suggestion.
If I have well understood I could use LIE to evaluate absolute DeltaG and rank my series of compounds.
But my problem is that almost the compounds I am testing are weak binders and the time of residence in the binding site during the MD simulation is of the order of 10 ns. Therefore, I am wondering whether I need a stable trajectory to run LIE.
Please could you tell me how much accurate is this method in predicting DeltaG?
LIE gives you the protein-ligand binding free energy (deltaG) as the difference in "solvation" of the ligand by two environments: (i) protein and solution, and (ii) solution. When you compare these deltaG values for different ligands, you obtain relative protein-ligand binding free energies (deltadeltaG).
10-ns timescale is sufficient, providing that you run a set of MD simulations for each ligand, at least three; the more, the better. In fact, it is generally better to run mutliple short MD simulations than one long MD simulation, regardless of the stability of the protein-ligand complex.
As for the conceptual soundness and reliability of this approach, see the works by Johan Aqvist:
https://www.ncbi.nlm.nih.gov/pubmed
Aqvist[author] and "linear interaction energy"[title/abstract]
- Badges
- Science topic