Positive binding energy. Changing dielectric constant didn't help.

[Anna Talyzina]

Dear users,

I was calculating binding energy between protein and ligand using default g_mmpbsa .mdp files. Forcefield, water model etc was also just like in the tutorial. The energy turned out to be huge and positive, due to the big and positive polar solvation energy. The error is also high.
As it was suggested previously in this list, I tried higher dielectric constant -pdie 50 and dielectric constant, calculated with g_dipoles for my system. The result was almost the same. Are there any other suggestions to solve positive energy problem?

van der Waal energy = -57.613 +/- 22.040 kJ/mol

Electrostattic energy = 27.316 +/- 105.664 kJ/mol

Polar solvation energy = 307.272 +/- 144.735 kJ/mol

SASA energy = -9.145 +/- 1.939 kJ/mol

Binding energy = 267.830 +/- 73.823 kJ/mol

Thank you!
Sincerely,
Anna

[Ann Talyzina]

[g_mmpbsa Mailing List]

The standard deviation for van der Waal, electrostatic and polar solvation energies are large. It is caused by either large conformational changes  or broken molecule across periodic box.

Concerning positive binding energies, MM/PBSA comes with limitations that are extensively discussed in literature. Sometimes MM/PBSA is not suitable for a specific system.