Dear all,
I checked more calculations and I found out the Eelec from receptor just
stays the same when I changed internal dielec constant (indi in MM/PBSA)
from 1, 2, to 4.
As we can see that Evdw and Enp stays the same, *the averaged Eelec and EPB
are gradually scaled down* when applying higher internal dielec constant
(which makes sense).
*But during the decomposition calculation, EPB are scaled down, too; but
Eelec stays the same which in theory, should scale down like EPB does. *
Since EPB scales down when indi increases, so the change of indi parameters
definitely has been passed to &decomp module...I don't think i missed some
other parameters under &decomp So...what is wrong with the decomposed
Eelec...is this a bug or so?
Best,
Guqin
Below are all the DELTA energies on same trajectory, with same radii, but
different indi params. The corresponding summed values from per-residue
decomposition are pasted, too.
*Differences (Complex - Receptor - Ligand): indi = 1.0*
Energy Component Average Std. Dev. Std. Err. of
Mean
-------------------------------------------------------------------------------
BOND -0.0000 0.0000
0.0000
ANGLE 0.0000 0.0000
0.0000
DIHED -0.0000 0.0000
0.0000
*VDWAALS -37.0580 4.0991
0.0826*
*EEL -51.7803 10.1739
0.2049*
1-4 VDW 0.0000 0.0000
0.0000
1-4 EEL -0.0000 0.0000
0.0000
*EPB 67.4662 7.8608
0.1583*
*ENPOLAR -5.9455 0.5523
0.0111*
EDISPER 0.0000 0.0000
0.0000
DELTA G gas -88.8383 9.4987
0.1913
DELTA G solv 61.5207 7.9174
0.1595
*DELTA TOTAL -27.3176 3.7704
0.0759*
-------------------------------------------------------------------------------
*Summed per-residue DELTA Total energy decomposition (ideomp=1) indi=1.0*
* EvdW Eelec
EPB Total*
BAZ 100 -18.52900487 -25.89016349 32.56959026 -11.84957809
Receptor 1-99 -18.52890548 -25.89008235 34.89674158 -9.522246247
** every energy components added up nicely here when indi = 1.0*
*Differences (Complex - Receptor - Ligand): indi = 2.0*
Energy Component Average Std. Dev. Std. Err. of
Mean
-------------------------------------------------------------------------------
BOND -0.0000 0.0000
0.0000
ANGLE 0.0000 0.0000
0.0000
DIHED -0.0000 0.0000
0.0000
*VDWAALS -37.0580 4.0991
0.0826*
*EEL -25.8902 5.0870
0.1025*
1-4 VDW 0.0000 0.0000
0.0000
1-4 EEL -0.0000 0.0000
0.0000
*EPB 33.3956 3.8584
0.0777*
*ENPOLAR -5.9455 0.5523
0.0111*
EDISPER 0.0000 0.0000
0.0000
DELTA G gas -62.9482 5.2571
0.1059
DELTA G solv 27.4501 3.9319
0.0792
DELTA TOTAL -35.4981 3.4884
0.0703
-------------------------------------------------------------------------------
*Summed per-residue DELTA Total energy decomposition** (ideomp=1) indi=2.0*
EvdW Eelec
EPB Total
BAZ 100 -18.52900487 -25.89016349 16.16567 -28.2535
Receptor 1-99 -18.52890548 *-25.89008235* 17.22993 -27.1891
*Differences (Complex - Receptor - Ligand): indi = 4.0*
Energy Component Average Std. Dev. Std. Err. of
Mean
-------------------------------------------------------------------------------
BOND -0.0000 0.0000
0.0000
ANGLE 0.0000 0.0000
0.0000
DIHED -0.0000 0.0000
0.0000
*VDWAALS -37.0580 4.0991
0.0826*
*EEL -12.9451 2.5435
0.0512*
1-4 VDW 0.0000 0.0000
0.0000
1-4 EEL -0.0000 0.0000
0.0000
*EPB 16.3237 1.8550
0.0374*
*ENPOLAR -5.9455 0.5523
0.0111*
EDISPER 0.0000 0.0000
0.0000
DELTA G gas -50.0031 3.9687
0.0799
DELTA G solv 10.3781 1.9657
0.0396
*DELTA TOTAL -39.6249 3.8697
0.0779*
-------------------------------------------------------------------------------
*Summed per-residue DELTA Total energy decomposition** (ideomp=1) indi=4.0*
EvdW Eelec
EPB Total
BAZ 100 -18.52900487 -25.89016349 7.950175 -36.469
Receptor 1-99 -18.52890548 *-25.89008235* 8.373466 -36.0455
On Mon, Sep 4, 2017 at 8:13 PM, Guqin Shi <shi.293.osu.edu> wrote:
> Dear Amber users,
>
> I am performing MM/PBSA calculation with idecomp =1 per-residue
> decomposition option on my protein-ligand system. The ligand is a small
> organic molecule.
>
> Just to be curious, I summed up decomposed per-residue energies, then I
> found the Eelec from decomposed file doesn't add up to the number reported
> in the final averaged result file...
>
>
> Below is averaged final result file:
>
> Differences (Complex - Receptor - Ligand):
> Energy Component Average Std. Dev. Std. Err. of
> Mean
> ------------------------------------------------------------
> ------------------------------------------------------------
> --------------------------------------
> BOND -0.0000 0.0000
> 0.0000
> ANGLE 0.0000 0.0000
> 0.0000
> DIHED -0.0000 0.0000
> 0.0000
> *VDWAALS -37.0580 4.0991
> 0.0826*
> *EEL -12.9451 2.5435
> 0.0512*
> 1-4 VDW 0.0000 0.0000
> 0.0000
> 1-4 EEL -0.0000 0.0000
> 0.0000
> *EPB 15.8420 1.8276
> 0.0368*
> *ENPOLAR -5.8373 0.5390
> 0.0109*
> EDISPER 0.0000 0.0000
> 0.0000
>
> DELTA G gas -50.0031 3.9687
> 0.0799
> DELTA G solv 10.0047 1.9391
> 0.0391
>
> *DELTA TOTAL -39.9984 3.8605
> 0.0778*
> ------------------------------------------------------------
> ------------------------------------------------------------
> --------------------------------------
>
>
>
>
> Below is the decomposed file that I processed and summarized receptor
> residues (1-99) and lig (100):
> ------------------------------------------------------------
> ------------------------------------------------------------
> --------------------------------------
> DELTAS:
> Total Energy Decomposition:
> Residue Location Internal *van der Waals Electrostatic*
> *Polar Solvation
> Non-Polar Solv. TOTAL*
> *BAZ 100* L BAZ 1 0 0 0 *-18.52900487* 2.049551273 0.041281011 *-25.89016349
> *5.086963618 0.102459013 *7.385438945* 0.907196193 0.018272281* 0* 0 0
> *-37.03372941* 3.942699281 0.079411828
> *Receptor 1-99 * summed up *-18.52890548*
> *-25.89008235* *8.456529817*
> *-35.96245801*
>
> ------------------------------------------------------------
> ------------------------------------------------------------
> --------------------------------------
>
> I highlighted the energy components. It seems like the EvdW is decomposed
> with half to receptor and half to ligand. And I know non-polar is not
> decomposable in MMPBSA. It seems like the EPB could be added up, too...
> But then the Eelec part seems to be the double of the average....Maybe I
> am missing some of the background knowledge of decomposition here or the
> electrostatic energy calculation...? Or, I double counted Electrostatic
> contribution when I directly summing up receptor residues 1-99? But all the
> other components seem to add up well... Can anyone explain a little bit to
> me?
>
>
>
> Thanks for the help!!!
> -Guqin
>
>
>
> --
> Guqin SHI
> PhD Candidate in Medicinal Chemistry and Pharmacognosy
> College of Pharmacy
> The Ohio State University
> Columbus, OH, 43210
>
>
--
Guqin SHI
PhD Candidate in Medicinal Chemistry and Pharmacognosy
College of Pharmacy
The Ohio State University
Columbus, OH, 43210
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Sep 05 2017 - 09:00:06 PDT
