cutoff of desolvation free energy for a favorable hydration site?

[zhentg]

Dear developers,

Inside the Waterkit paper, a cutoff of -5 kcal/mol is used to select favorable Waterkit hydration site for comparison with GIST result.
Is -5 kcal/mol a recommended cutoff to define a favorable hydration site, any data to support this?

What is the range of the desolvation free energy(DFE)? For the paper and my tests, DFE is always below 0, is it correct?

Best
Zhenting

[jeeberhardt]

Hi Zhenting,

The cutoff of -5 kcal/mol was defined more or less arbitrarily based on the number of identified hydration sites with a dG below -5 kcal/mol, considering all the systems together. So we just wanted to have enough data to make the analysis meaningful.

Considering the range, it is highly variable and depends of your system. Based on the results (see figures S35 to S44 in the supplementary data), the dG can go from -12 to 0 kcal/mol. But this value also depends on the water_radius you choose (default 1.4 A). A bigger radius should makes dG more favorable (as you sum over a bigger number of GIST grid points), while a smaller one will produce less favorable dG.

Additionally, GIST maps can be combined differently, see this paper (and the supplementary materials): Testing inhomogeneous solvation theory in structure-based ligand discovery (2017)

Best,
Jerome.

[zhentg]

Thanks a lot, Jerome. The reference you suggest is helpful. I believe incorporation of IST into docking would be valuable to improve the docking pose and affinity prediction.

On the other hand, about the cutoff and range of dG, do you know if by definition, GIST dG will be <=0 kcal/mol? Schrodinger WaterMap is also based on GIST, but its dG ranges from -10~10 kcal/mol.