[gmx-users] Temperature and pressure scaling with frozen protein

[Michel Cuendet]

Hi Folks,

  I'm trying to equilibrate a big protein (135x75x75 box). The structure 
is quite fragile, so it won't stand the initial pressure scaling. I have 
to freeze it first, equilibrate the water, and then do an energy 
minimization on the whole system, and heat up slowly.

  I observe a constant temperature increase during the first 
equilibration run (with protein and crystal water frozen), reaching more 
than 320K after 100ps. The increase continues after the box has 
stabilized. How is this possible with a thermostat? The parameters I use 
are given below.

  I saw this subject had already benn discussed in the list, but 
couldn't figure out what the conclusions were. What should I do in my 
case, where I have to do pressure scaling with a frozen group?

  Thanks,

Michel Cuendet.

___________________________________________________________________

integrator               = md
dt                       = 0.001
nsteps                   = 50000
comm_mode         = Linear
nstcomm             = 1
comm_grps         = Protein

tcoupl                   = berendsen
tc-grps                  = FreeWAT Protein_CrystalWater
tau_t                    = 0.1  0.0
ref_t                    = 300.0  0.0
Pcoupl                   = berendsen
tau_p                    = 1.0
compressibility          = 4.5e-5
ref_p                    = 1.0

energygrps        = Protein_CrystalWater
freezegrps        = Protein_CrystalWater
freezedim        = Y Y Y
energygrp_excl          = Protein_CrystalWater Protein_CrystalWater