[gmx-users] problems about energy minimization

[Mark Abraham]

renying wrote:
> Hello, everyone.
> I have loaded a pdb file of protein complex from Protein Data Bank and put it into water soltion with pdb2gmx, editconf and genbox.
> After 400 steps of energy minimization when the potential energy seems to be more or less stable, I want to do position restraints. However, it always failed. 
> 
> The error info is as follows:
> Step 1, time 0.002 (ps)  LINCS WARNING
> relative constraint deviation after LINCS:
> max 0.000645 (between atoms 13241 and 13242) rms 0.000043
> bonds that rotated more than 30 degrees:
>  atom 1 atom 2  angle  previous, current, constraint length
>   13241  13242   31.2    0.1000   0.0999      0.1000
> 
> t = 0.002 ps: Water molecule starting at atom 73326 can not be settled.
> Check for bad contacts and/or reduce the timestep.Wrote pdb files with previous and current coordinates
> Large VCM(group System):      0.03330,     -0.01095,     -0.00024, ekin-cm:  2.20013e+03

Check out http://wiki.gromacs.org/index.php/Errors#LINCS_warnings for
some ideas. Looking at this water molecule's environment can help.

> The system seems to be exploding. I have tried the follwing parameters and sadly, none of them works.
> nsteps              =  200/400/800/1000 steps  in em.mdp
> dt                  =  0.002/0.001/0.0002 ps  in pr.mdp
> tc-grps             =  Protein  SOL  or    
> tc-grps             =  system         in pr.mdp
> tau_t               =  0.002    0.002  or longer time         in pr.mdp

If you're sure the problem isn't a water trapped inside a protein or
suchlike, one thing you can try is a slow heating. You'll need to do a
series of .mdp files with an initial gen_vel to (e.g.) 50K, and then use
 ref_t to increase that in similar increments.

> The total charge of the system is  -273.000 e and I didn't neutralize it for some reasons. Does this matter much?

It probably isn't breaking the numerical simulation, but it could well
mean you're not modelling reality. Nature abhors a vacuum, and an
isolated charge :-)

Mark