[gmx-users] Energy mimimization in AcOH box

[Aditi Borkar]

Dear All,

Hello!

I am simulating a protein in a box of 9M acetic acid. I have obtained
the coordinates and gromacs topology for AcOH from PRODRG. The AcOH
has net -1 charge and the topology file is as follows:

[ moleculetype ]
; Name nrexcl
ACA      3

[ atoms ]
;   nr      type  resnr resid  atom  cgnr   charge     mass
     1       CH3     1 ACA     CAA     1    0.056  15.0350
     2         C       1  ACA     CAD     1    0.393  12.0110
     3        OM     1  ACA     OAC     1   -0.725  15.9994
     4        OM     1  ACA     OAB     1   -0.724  15.9994

[ bonds ]
; ai  aj  fu    c0, c1, ...
   1   2   1    0.153    334720.0    0.153    334720.0 ;   CAA  CAD
   2   3   1    0.125    418400.0    0.125    418400.0 ;   CAD  OAC
   2   4   1    0.125    418400.0    0.125    418400.0 ;   CAD  OAB

[ pairs ]
; ai  aj  fu    c0, c1, ...

[ angles ]
; ai  aj  ak  fu    c0, c1, ...
   1   2   3   1    120.0       418.4    120.0       418.4 ;   CAA  CAD  OAC
   1   2   4   1    120.0       418.4    120.0       418.4 ;   CAA  CAD  OAB
   3   2   4   1    126.0       502.1    126.0       502.1 ;   OAC  CAD  OAB

[ dihedrals ]
; ai  aj  ak  al  fu    c0, c1, m, ...
   2   1   4   3   2      0.0 1673.6        0.0 1673.6   ; imp   CAD
CAA  OAB  OAC

When I had tried taking uncharged AcOH, during equilibration, the
water and AcOH were completely segregating into two layers (like a
biphasic system).

My problem arises when I perform energy mimization on my solvated
protein. With 300 steps of EM (emtol 1000 and emstep .01) the EM
doesn't converge and more distrurbingly within these few steps, the
secondary structure of my protein is completely lost. This problem
does not occur when I do EM in box containing protein and pure water.
The Fmax is displayed onto 2 TRP residues.

My em.mdp file is as follows:

define           =  -DFLEXIBLE
constraints      =  none
integrator       =  steep
dt               =  0.002    ; ps !
nsteps           =  300
nstlist          =  10
ns_type          =  grid
rlist            =  0.9
coulombtype      =  PME
rcoulomb         =  0.9
rvdw             =  0.9
fourierspacing       = 0.12
fourier_nx     =  0
fourier_ny     =  0
fourier_nz     =  0
pme_order      =  4
ewald_rtol     =  1e-5
optimize_fft         =  yes
;
;      Energy minimizing stuff
;
emtol              =  1000.0
emstep             =  0.01

Please help!

Thank you!

-Aditi Borkar,
Tata Institute of Fundamental Research,
Mumbai,
India.