[gmx-users] Re: Protein ligand simulation

[Mark Abraham]

On 2/12/2011 12:37 PM, bharat gupta wrote:
> Hi,
>
> I am trying the simulation of a docked complex of my protein . While 
> solvating the box using the following command :-
> genbox -cp newbox.gro -cs spc216.gro -p topol.top -o solv.gro
>
>
> The processing does not stop and continues to run . Here's the output 
> that I got while solvating the box , which is still continuing :-
>
>
> Reading solute configuration
> GRoups of Organic Molecules in ACtion for Science
> Containing 3620 atoms in 231 residues
> Initialising van der waals distances...
>
> WARNING: masses and atomic (Van der Waals) radii will be determined
>          based on residue and atom names. These numbers can deviate
>          from the correct mass and radius of the atom type.
>
> Reading solvent configuration
> "216H2O,WATJP01,SPC216,SPC-MODEL,300K,BOX(M)=1.86206NM,WFVG,MAR. 1984"
> solvent configuration contains 648 atoms in 216 residues
>
> Initialising van der waals distances...
> Will generate new solvent configuration of 46x46x46 boxes
> Generating configuration
> Sorting configuration
> Found 1 molecule type:
>     SOL (   3 atoms): 21024576 residues

This is 21 million water molecules, from far too many boxes for a 3K 
atom protein. A few tens of thousands of water molecules would be 
normal. Something is wrong with the coordinates or box dimensions of 
your -cp file.

> Calculating Overlap...
> box_margin = 0.315
> Removed 1047843 atoms that were outside the box
>
> What have gone wrong for solvation to take this much time on 12 
> processors CPU.

All water atoms have to be tested for overlap with all protein atoms - 
that's expensive. Also, no GROMACS tools apart from mdrun actually use 
more than one processor of those available.

Mark