positive g_mmpbsa binding energy

[Sajad Ahrari]

Dear Users,

I used the attached pbsa.mdp file with the "g_mmpbsa -s ../../5es1_100ns_nowater.tpr -f  5es1_5RC_last5ns_dt80.xtc  -i pbsa.mdp -n ../../../MD_preparation/index.ndx -pdie 10 -pbsa -decomp " command to calculate binding free energy between a small molecule and a protein. Although the ligand has a nanomolar IC50 for the protein, but the binding energy is positive according to g_mmpbsa calculations. I increased the dielectric constant by increasing the pdie to 10. However, the final calculated free energy was still positive. Could you please help me know if I am missing something and if it is normal to get such results?

Thank you,

Sajjad

[g_mmpbsa Mailing List]

It is very difficult to say, why binding energy is positive. It appears that polar solvation energy is not compensated by electrostatic energy. Have you used a crystal structure or docked complex? In case of docked complexes energy can be positive as docked pose might not be favorable for binding.

Why you have used dielectric constant 20? Larger you take this constant, lower will be electrostatic energy and also affect polar solvation energy.

[Sajad Ahrari]

I used a crystal structure and started with 2 as the dielectric constant and the binding energy was still positive. I increased the dielectric constant, based on an advise I found on the web (http://gromacs.org_gmx-users.maillist.sys.kth.narkive.com/mKjZwTIN/g-mmpbsa-positive-binding-energy). Do you think I should reduce the dielectric constant? As I measured, the real dielectric constant should be around 1.

Thank you.

[g_mmpbsa Mailing List]

What force-field you are using for the simulations?

[Sajad Ahrari]

I am using charmm-36.ff. And for the ligand parameters I used the "https://cgenff.paramchem.org/initguess/".