CHARMM Development Project Forums User Discussion & Questions Energy terms, Constraints, Restraints, and Solvation solvated energy calculation with gbmv

What I want to do is calculating the solvated energy of DNA by gbmv approach. All I can find about this topic is the test script in charmm. I used the second analytical method. Is the parameter setting good enough to accomplish the task? Thanks a lot!


!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
faster on
update cutnb 999 ctofnb 998 ctonnb 997 eps 1 vswi switch


!************************************
! Set version # to PARAM19, parameters were fitted to PARAM22.
! However, errors are pretty small < 1.5 % for PARAM19.
! Errors are ~ 1 % for PARAM22 vs. 0.25 A - PB Calculations
!************************************

set ver = 19
if ver .eq. 22 set GBPARAMS = SHIFT -0.102 SLOPE 0.9085
if ver .eq. 19 set GBPARAMS = SHIFT -0.102 SLOPE 0.9085



!***********************************************
! Start GBMV, Method II: Recommended parameters
! Non-zero SA term turns on surface area.
! By default, no SA term is calculated.
!***********************************************

GBMV BETA -20 EPSILON 80 DN 1.0 watr 1.4 GEOM -
TOL 1e-8 BUFR 0.5 Mem 10 CUTA 20 HSX1 -0.125 HSX2 0.25 -
ALFRQ 1 EMP 1.5 P4 0.0 P6 8.0 P3 0.70 ONX 1.9 OFFX 2.1 -
SON 0.5 SOFF 1.75 -
WTYP 2 NPHI 38 CORR 1 SA 0.00592 SB 0.0 @GBPARAMS

!************************************
! Simple energy evaluation and alpha show
!************************************
energy

scalar gbalpha show

!************************************
! Minimization should go downhill
!************************************

mini abnr nstep 1000

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

param19 for DNA??
You could compare with a Poisson-Boltzmann calculation.

I'm using param22 force field for the calculation. The command doesn't matter because the variable GBPARAMS is the same for both cases.

Yes, PARAM19 is for proteins only; there is an analogous extended (united) atom set for nucleic acids (toprna10r_22.inp, pardna10_22.inp) but I'm not sure if solvation parameters for it have been included in any of GB based implicit solvent models.

Note that param22 is also for proteins only; the nucleic acid set is param27.

My charmm is a little bit old (c32b1), so I'm not sure whether the force field I'm using is param22 or param27. But it does have residue topology definition for RNA as well as patches for DNA. When did param27 come out?

Since c22, the parameter set version numbers roughly correspond to the CHARMM release when they were first included in a public 'beta' distribution. Updated and more recent parameters are available via the MacKerell Lab; for carbohydrate, ether, and the new general small molecule force fields you would need to update your CHARMM release, however.

I compared the solvation energies from gbmv with those from pbeq (grid size=0.4). The error is about 1%. Pretty close!

gbmv pbeq
-4573.3 -4536
-4586.8 -4545.6
-4567.6 -4536.6
-4591.4 -4542.5

Nice. What's the speed difference?