[gmx-users] RE:Range checking error

[Justin A. Lemkul]

lloyd riggs wrote:
> Dear All,
> 
> Before I start, I have already searched through the mailing list archives, etc...and have energy minimized my structure (s) ;
> 
> I get the range check error;
> _________________________________________________________________________
> Range checking error:
> Explanation: During neighborsearching, we assign each particle to a grid
> based on its coordinates. If your system contains collisions or parameter
> errors that give particles very high velocities you might end up with some
> coordinates being +-Infinity or NaN (not-a-number). Obviously, we cannot
> put these on a grid, so this is usually where we detect those errors.
> Make sure your system is properly energy-minimized and that the potential
> energy seems reasonable before trying again.
> 
> Variable ci has value -2147483648. It should have been within [ 0 .. 19683 ]
> __________________________________________________________________________
> 
> does anyone have a fix for this?  
> 
> I have first energy minimized my structures.  There are two seperate proteins with two seperate protein domains (chain a,b each) making up two receptors which interact with eachother transiently, or strongly in the presence of a molecule.  The two proteins are seperated (manually) by 10 angstroms.
> 
> I first energy minimized these, then added waters and ions.  I had problems with force fields that recognized all the ions (Mg2+, Ca2+, K+, Na+, Cl-) and ended up using the oplsaa force fields.
> 
> Now, I try an initial run, MD or a simple 100 step energy minimization, prior to a long run, and get the above error again and again.  The two scripts mentioned are below inline.  I have however played around with everything I could.  I have also started from scratch 3 times, ie the initial proteins-minimize-add waters/ions-and then it crashes?
> 
> Also, does anyone know what the variable "ci" is?  
> 

<snip>

> integrator = md 
> dt = 0.005 

With a timestep of 5 fs, you should (at minimum) be using constraints, and 
probably virtual sites as well.  Does a more sensible timestep (like 1 or 2 fs) 
make your system stable?

> nsteps = 1000 
> nstxout = 100000 
> nstvout = 100000 
> nstlog = 1000 
> nstenergy = 1000 
> nstxtcout = 1000 
> xtc_grps = system 
> energygrps = system 
> nstlist = 20 

You should aim for an nstlist that will update your neighborlist roughly every 
10-20 ps, or else (in theory) you could be missing interactions.  With your 
neighborlist updated every 100 ps (!) then your short-range interactions could 
be changing quite a bit.

> coulombtype = PME 
> ns_type = grid 
> rlist = 1.5 
> rcoulomb = 1.5 
> rvdw = 1.5 
> pbc = xyz 
> table-extension = 20
> 
> ;Temperature coupling 
> tcoupl  = berendsen
> tc-grps = system 

Coupling the whole system in one temperature bath may or may not be appropriate. 
  Although not the most likely culprit for your problem, you should be aware. 
See here, particularly the cited references:

http://www.gromacs.org/Documentation/Terminology/Thermostats

> tau_t = 1.0
> ref_t = 50.0 
> 
> ; Isotropic pressure coupling is now on
> Pcoupl              =  berendsen
> Pcoupltype          =  isotropic
> tau_p               =  2.01
> compressibility     =  4.5e-5
> ref_p               =  1.0
> refcoord_scaling    =  all
> 
> ;Velocity generation 
> gen_vel = yes 
> gen_temp = 50.0 
> gen_seed = 30 
> ld_seed = 130 
> comm_mode = angular 
> 

In a periodic system, using this comm_mode is inappropriate.  Surely grompp 
warned you about this?

-Justin


-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================