Question about MMGBSA Energy Decomposition in MOLAICAL

[AhmadAlqaisi]

Dear Dr. Qifeng Bai, I work in computational chemistry. I have been using the MOLAICAL code for MMGBSA calculations for NAMD 100ns trajectories, and it has been working greatly, specially with the new updated version 1.1 I have a question about the MMGBSA result from MOLAICAL. Can we determine each energy type by itself? the delta E(internal) is stated by itself, however delta E(electrostatic) and delta E(VDW) are not very clear.  Now I assumed that I can find the electrostatic and vdW energies by using: Elec(complex) - (Elec(pro)+Elec(lig)) vdW(complex) - (vdW(pro)+vdW(lig)) and in this way, I can find deltaG(sol). I expected deltaG(sol) to be the same from both equations but it's not. So, I'm assuming that deltaG(sol) in the case of the electrostatic energy is the polar contribution and is the non-polar contribution in the case of vdW. Can you confirm if my understanding is correct or not? I also included a MMGBSA result if you want to demonstrate anything  Looking forward to your reply Best regards,Ahmad Alqaisi

" Total run 1000 frames from complex.log. Average BOND: 671.7435284999997 Average ANGLE: 1717.4360980000001 Average DIHED: 1908.5886183999967 Average IMPRP: 102.70511910000002 Average ELECT: -5919.676528199999 Average VDW: -602.7139261000007 ###############################

Total run 1000 frames from protein.log. Average BOND: 647.4863226000006 Average ANGLE: 1662.567088499999 Average DIHED: 1863.0686601000007 Average IMPRP: 101.58399230000015 Average ELECT: -5809.532617400003 Average VDW: -608.4611837999997 ###############################

Total run 1000 frames from ligand.log. Average BOND: 24.25720030000002 Average ANGLE: 54.869011799999946 Average DIHED: 45.51996180000004 Average IMPRP: 1.1211298000000014 Average ELECT: -113.74073879999987 Average VDW: 31.277075 ###############################

delta E(internal): -3.200003959591413E-6 delta E(electrostatic) + deltaG(sol): 3.5968280000040096 delta E(VDW) + deltaG(sol): -25.529817300000932 delta G binding: -21.932992500000882  +/-  0.366 (kcal/mol) "

[MolAICal]

Dear Dr. Qifeng Bai, I work in computational chemistry. I have been using the MOLAICAL code for MMGBSA calculations for NAMD 100ns trajectories, and it has been working greatly, specially with the new updated version 1.1 I have a question about the MMGBSA result from MOLAICAL. Can we determine each energy type by itself? the delta E(internal) is stated by itself, however delta E(electrostatic) and delta E(VDW) are not very clear.  Now I assumed that I can find the electrostatic and vdW energies by using: Elec(complex) - (Elec(pro)+Elec(lig)) vdW(complex) - (vdW(pro)+vdW(lig)) and in this way, I can find deltaG(sol). I expected deltaG(sol) to be the same from both equations but it's not. So, I'm assuming that deltaG(sol) in the case of the electrostatic energy is the polar contribution and is the non-polar contribution in the case of vdW. Can you confirm if my understanding is correct or not? I also included a MMGBSA result if you want to demonstrate anything  Looking forward to your reply Best regards,Ahmad Alqaisi

" Total run 1000 frames from complex.log. Average BOND: 671.7435284999997 Average ANGLE: 1717.4360980000001 Average DIHED: 1908.5886183999967 Average IMPRP: 102.70511910000002 Average ELECT: -5919.676528199999 Average VDW: -602.7139261000007 ###############################

Dear Ahmad Alqaisi,

Yes, you are right.

Best, Qifeng

[MolAICal]

I am sorry for removing item due to operate my pad wrongly. I paste the replied issue of Ahmad Alqaisi as below:

Hello Dr. Qifeng,

Thank you for your reply. To be honest, I'm still confused and didn't
understand your response in an accurate manner. In my scenario, I'm doing
MMGBSA binding free energy for a protein-ligand complex. After the
calculation, I get the output that I have shown in the previous email.

In the output shown of the MOLAICAL MMGBSA calculation for the
protein-ligand complex binding energy, can I get the solvation polar (Gpol)
and nonpolar (Gnon) contributions of the binding free energy ? as well as
the Eelec and EvdW individually?

The way that I have done that is:

*If delta **E(electrostatic)* *+ deltaG(sol) = 3.5968280000040096*

and If I calculate the *delta **E(electrostatic)* from taking the *Average
ELECT* value from the sections (complex, ligand, protein) by

*delta **E(electrostatic) = **Average ELECT**(complex)-(**Average ELECT*
*(protein)+**Average ELECT**(ligand))*

then

*delta G(sol) = 3.5968280000040096 - E(electrostatic)*

and If I wanted the non-polar, I should do the same thing but for vdW. Is
this procedure correct?

Best regards,
Ahmad Alqaisi

[MolAICal]

Dear Ahmad Alqaisi,

I am sorry for misunderstanding your above question.

Please check new tutorial for mm/gbsa in MolAICal site. VDW can be got independently.

Your formulae are correctly. But, actually, the ELECT column of the NAMD log file consists of total electrostatic contribution including ∆G(solv). Currently, I do not find the suitable way to decompose total electrostatic contribution and ∆G(solv).

Please read this paper: https://pubs.acs.org/doi/10.1021/ja00172a038

∆G(solv) = ∆G(cav) + ∆G(vdw) + ∆G(pol).

It is not very easy to decompose them.

best, qifeng