hi<br>> I m using gromacs4.5.3 . hardware 1GB ram 320 hard disk .<br>when i skiped scaling by factor 0.95 then , after EM step I m getting same potential energy nan result.<br><br><br><br><br><br><br><br><br>> i m doing simulation membrane protein<br>
> while scaling down the lipids by a factor of 0.95 then performing EM<br>><br>> Steepest Descents converged to Fmax < 10 in 18 steps<br>> Potential Energy = -nan<br>> Maximum force = 4.7791553e+02 on atom 5440<br>
> Norm of force = -nan<br>><br>><br>><br>> ; minim.mdp - used as input into grompp to generate em.tpr<br>> ; Parameters describing what to do, when to stop and what to save<br>> integrator = steep ; Algorithm (steep = steepest descent<br>
> minimization)<br>> emtol = 10.0 ; Stop minimization when the maximum force <<br>> 1000.0 kJ/mol/nm<br>> emstep = 0.1 ; Energy step size<br>> nsteps = 25000 ; Maximum number of (minimization) steps<br>
> to perform<br>><br>> ; Parameters describing how to find the neighbors of each atom and how<br>> to calculate the interactions<br>> nstlist = 10 ; Frequency to update the neighbor list and<br>
> long range forces<br>> ns_type = grid ; Method to determine neighbor list<br>> (simple, grid)<br>> rlist = 1.2 ; Cut-off for making neighbor list (short<br>> range forces)<br>> coulombtype = Shift ; Treatment of long range electrostatic<br>
> interactions<br>> rcoulomb = 1.2 ; Short-range electrostatic cut-off<br>> rvdw = 1.2 ; Short-range Van der Waals cut-off<br>> pbc = xyz ; Periodic Boundary Conditions (yes/no)<br>
> define = -DSTRONG_POSRES<br>><br>><br>> earlier when i was using<br>><br>> ; minim.mdp - used as input into grompp to generate em.tpr<br>> ; Parameters describing what to do, when to stop and what to save<br>
> integrator = steep ; Algorithm (steep = steepest descent<br>> minimization)<br>> emtol = 1000.0 ; Stop minimization when the maximum force <<br>> 1000.0 kJ/mol/nm<br>> emstep = 0.01 ; Energy step size<br>
> nsteps = 50000 ; Maximum number of (minimization) steps<br>> to perform<br>><br>> ; Parameters describing how to find the neighbors of each atom and how<br>> to calculate the interactions<br>
> nstlist = 1 ; Frequency to update the neighbor list and<br>> long range forces<br>> ns_type = grid ; Method to determine neighbor list<br>> (simple, grid)<br>> rlist = 1.2 ; Cut-off for making neighbor list (short<br>
> range forces)<br>> coulombtype = PME ; Treatment of long range electrostatic<br>> interactions<br>> rcoulomb = 1.2 ; Short-range electrostatic cut-off<br>> rvdw = 1.2 ; Short-range Van der Waals cut-off<br>
> pbc = xyz ; Periodic Boundary Conditions (yes/no)<br>> define = -DFLEXIBLE<br>><br>><br>> then problem was same<br>><br><br>Which version of Gromacs? On what hardware? I discovered a platform-specific<br>
bug that looked a lot like this, but I hesitate to suggest that until I know more.<br><br>In all likelihood, you simply have unresolvable atomic overlap (i.e., you're<br>packing too much) such that energy minimization cannot complete, since nan =<br>
"not a number," or something is infinitely large or small.<br><br>