Re: RATTLE algorithm error

From: Peter Freddolino (petefred_at_ks.uiuc.edu)
Date: Wed Apr 11 2007 - 19:09:03 CDT

Hi Marcelo,
you might want to check a couple things:
-Is it always the same atom that causes a constraint failure? Looking at
the place where the failure occurs, and the immediate environment, can
help determine the cause
-Does the failure occur at or close to one of your constrained atoms?
You're using a very sharp constraint function (r^6), which may not work
very well with the timestep you're using since if an atom moves too far
in one step it will experience a very large constraint force.
-How quickly does the failure occur?

Peter

Marcelo Puiatti wrote:
> Dear users:
>
> I'm trying to run a NTV equilibration. But I allways find the same
> message.
>
> ERROR: Constraint failure in RATTLE algorithm for atom nnnn!
>
> I tried to find the solution to my problem in the list, I found a lot
> off scenarios where the same problem appears and in VERY DIFFERENT
> condittions. However, it seems there are to possible solutions
> (according to the list):
>
> Modiffing the PME grid size
>
> Relaxing the structure more steps
>
> Which is the best solution?
> Is this problem directly related with my structure?
> I work with the same structure with AMBER and it does work nicelly
>
> Is there any cuttoff in the Gradient that make things don't work?
>
>
> Thanks in advance
>
>
> Marcelo
>
> P.D. The protein contains 1120 residues and aprox. 28200 solvent molecules
> I attached the config file:
>
> #############################################################
> ## ADJUSTABLE PARAMETERS ##
> #############################################################
>
> set temperature 300
> set outputname NTV
>
> firsttimestep 0
>
> #############################################################
> ## SIMULATION PARAMETERS ##
> #############################################################
>
> # Input
> paraTypeCharmm off
> paratypeXplor off
> amber on
> parmfile myprotein.top
> ambercoor myprotein.restrt
> # parameters ../common/par_all27_prot_lipid.inp
> temperature $temperature
>
>
> # Force-Field Parameters
> exclude scaled1-4
> 1-4scaling 0.833333 # =1/1.2, default is 1.0
> cutoff 11.5
> switching on
> switchdist 9.
> pairlistdist 13.5
>
>
> # Integrator Parameters
> timestep 1.5 ;# 2fs/step
> rigidBonds all ;# needed for 2fs steps
> nonbondedFreq 1
> fullElectFrequency 2
> stepspercycle 20
>
>
> # Constant Temperature Control
> langevin on ;# do langevin dynamics
> langevinDamping 5 ;# damping coefficient (gamma) of 5/ps
> langevinTemp $temperature
> langevinHydrogen off ;# don't couple langevin bath to hydrogens
>
>
> # Periodic Boundary Conditions
> cellBasisVector1 88.15 0. 0.
> cellBasisVector2 0. 88.72 0.
> cellBasisVector3 0. 0.0 153.14
> cellOrigin 45.52 45.94 78.17
>
> wrapAll on
>
>
> # PME (for full-system periodic electrostatics)
> PME yes
> PMEGridSizeX 128
> PMEGridSizeY 128
> PMEGridSizeZ 194
>
>
> # Constant Pressure Control (variable volume)
> useGroupPressure yes ;# needed for rigidBonds
> useFlexibleCell no
> useConstantArea no
>
> #Constraints
> constraints on
> consexp 6
> consRef myproteinConst.pdb
> consKFile myproteinConst.pdb
> consKCol B
>
> # Output
> outputName $outputname
>
> restartfreq 200 ;# 500steps = every 1ps
> dcdfreq 500
> xstFreq 500
> outputEnergies 50
> outputPressure 50
>
>
> #############################################################
> ## EXTRA PARAMETERS ##
> #############################################################
>
>
> #############################################################
> ## EXECUTION SCRIPT ##
> #############################################################
>
> # Minimization
> # minimize 2000
> reinitvels $temperature
>
> run 50000
>
>
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:46:11 CST