[gmx-developers] distance_restraints for shell particle

[Leontyev Igor]

Hi,
I would like to restrain a distance of shell particles. Is it possible for 
the massless particle? My system is as simple as a box 3-site water 
molecules with one additional shell particle per each molecule. Parallel 
simulations does not start due to the following error:
-------------------------------------------------------
Program mdrun, VERSION 4.0.5
Source code file: disre.c, line: 135

Fatal error:
Time or ensemble averaged or multiple pair distance restraints do not work 
(yet) with domain decomposition, use particle decomposition (mdrun 
option -pd)
------------------------------------------------------------

But I do neither time averaging nor multiple pair distance restraints. Once 
I switched on single-cpu mode the error is:

-------------------------------------------------------------

           Step           Time         Lambda
              0        0.00000        0.00000

RELAX: Using prediction for initial shell placement
Grid: 4 x 4 x 4 cells
Long Range LJ corr.: <C6> 1.6365e-04
Long Range LJ corr.: Epot   -45.4211, Pres:    -82.357, Vir:    45.4211
   Energies (kJ/mol)
        LJ (SR)  Disper. corr.   Coulomb (SR)   Coul. recip.   Polarization
    4.17424e+07   -4.54211e+01   -3.64965e+04   -1.50620e+03    1.31274e+04
     Dis. Rest. D.R.Viol. (nm)      Potential    Kinetic En.   Total Energy
    0.00000e+00    0.00000e+00    4.17175e+07    2.28615e+07    6.45790e+07
  Conserved En.    Temperature Pressure (bar)
    6.45790e+07    1.79185e+06    1.64771e+08

t = 0.000 ps: Water molecule starting at atom 2001 can not be settled.
Check for bad contacts and/or reduce the timestep.
-------------------------------------------------------------------

There is no bad contacts for sure. The initial structure was minimized with 
SPC/E  topology then positions of shell particles were set up equal to those 
of corresponding water oxygen.  Moreover, the simulation starts with no 
problems once I turn off the distance restraints (disre=no). Any thoughts? 
Thanks.
Bellow is the topology:



[ atomtypes ]
;name  at.num     mass      charge   ptype         sigma         eps 
[kj/mol]
   OW    8      15.99940       0.000       A   0.31660e+00   0.65000e+00
   HW    1       1.00800       0.000       A   0.00000e+00   0.00000e+00
   DW    0       0.00000       0.000       S   0.00000e+00   0.00000e+00

[ moleculetype ]
; molname nrexcl
SOL  2

[ atoms ]
; id   at type res nr  residu name    at name cg nr     charge
  1      OW      1      SOL            OW      1       1.5000
  2      HW      1      SOL            HW1     1       0.3000
  3      HW      1      SOL            HW2     1       0.3000
  4      DW      1      SOL            DW      1      -2.1000

[ polarization ]
;ai     aj     type     alpha(nm^3)
 1 4 1       1.47e-03

[ distance_restraints ]
;ai     aj     type     index      type'     low   up1   up2    fac
  1      4       1        0         1        0.0   0.05  0.4     1

[ settles ]
; OW    funct   doh        dhh
1       1       0.1000    0.16330

[ bonds ]
    1     4     5
    2     4    5
    3     4     5

[ exclusions ]
1 2 3
2 1 3
3 1 2