Hi,
From the MD module documentation, it says "Long range electrostatic interactions are implemented using the smooth particle mesh Ewald technique, for neutral periodic cubic systems in the constant volume ensemble, using pair interaction potentials". Does this mean this method only applies for the exact cubic boxes and not any rectangular boxes?? I have tried this option for my system in the rectangular box and have seen the output for electrostatic energy is too way off from the case with 'PME off'. My box size is 57 x 66 x 97 A^3 and my md input directive is
start dutch
md
system dutch_md
update pairs 10
cutoff 1.2
pme grid 16
noshake solute
isotherm 0 298.15 anneal 0 5 trelax 0.1 0.1
equil 0
time 0.0
step 0.001
data 25000
print step 100 stat 1000
record rest 100 prop 100 coord 100
mwm 4000
end
task md dynamics
The U(ele) with PME 'on' is about -3.0814E+06 kj/mol while without PME is about -6.7318E+05 kj/mol. The electrostatic energy of the latter case is comparable to that obtained from other MD package such as NAMD. The weird thing is the U (tot) for both cases are in the range of 10^5 kj/mol, which doesn't look reasonable for the first case that has large electrostatic contribution.
So I'm wondering whether the PME doesn't work for the system other than periodic cubic box or my setup is missing something.
Best,
Tee
|