covalent docking free energy estimate

[agoliaei]

Hi, I think I managed to figure out how to run covalent docking using both Meeko and AD-GPU. I am wondering, is it expected to see a large positive number as the estimated free energy, or am I making a mistake the way I set up my simulation? I guess I am confused on the energy terms that are used for "Final Total Internal Energy".

DOCKED: USER Estimated Free Energy of Binding = +66.49 kcal/mol [=(1)+(2)+(3)-(4)] DOCKED: USER DOCKED: USER (1) Final Intermolecular Energy = +0.00 kcal/mol DOCKED: USER vdW + Hbond + desolv Energy = +0.00 kcal/mol DOCKED: USER Electrostatic Energy = +0.00 kcal/mol DOCKED: USER Moving Ligand-Fixed Receptor = +0.00 kcal/mol DOCKED: USER Moving Ligand-Moving Receptor = +0.00 kcal/mol DOCKED: USER (2) Final Total Internal Energy = +66.49 kcal/mol DOCKED: USER (3) Torsional Free Energy = +0.00 kcal/mol DOCKED: USER (4) Unbound System's Energy = +0.00 kcal/mol

[diogomart]

Hi, can you share your input files and the AD-GPU command you used?

[agoliaei]

Upon further exploring, looks like AD4 handles covalent binding in a different way than AD-GPU. The estimated free energy looks more realistic, however, it takes forever to finish calculations on CPU. I have a question, looks like in covalent binding all the "inter" molecular terms are set to zero. Does this mean docking a ligand that is bound covalently is not taking vdW, Hbond, Elect,... energies between the ligand and the rest of the protein into account? How can this result in finding a reliable pose (redocking experiments)? Am I understanding this correctly?

[agoliaei]

Closing as I found the problem. My set up was incorrect. Thank you!