One thing you could check is take the last restart file, and run your
simulation for another 10,000 steps or so in the NVE ensemble. Check to see
if there is substantial energy drift. If so, reduce the timestep. Also, if
you are using shake, make sure you have to constraints specified correctly
Regards,
Conor Parks
B.S.E in Chemical Engineering, University of Michigan, 2012
PhD candidate in Chemical Engineering, Purdue University
On Fri, Mar 11, 2016 at 8:41 PM, Arati Paudyal <apsilwal123.gmail.com>
wrote:
> Dear all,
>
> I just finished a 1 ns heating and a 60 ns production simulation. But when
> I processed the RMSD I feel like it is too high. I know the fluctuation
> from 1-3 angstron is expected but mine looks like it is over 4 sometimes.
> Could you please explain what might have gone wrong during simulation? IS
> there anything I can do to avoid such fluctuation during simulation? I
> used exactly same input and time to run another simulation with a different
> protein complex and rmsd was a lot lower in that case. I know it depends on
> the system but if I could understand if there is anything I can do to
> improve this it would be of great help. Can I even use any of these frames
> to process mmpbsa script?
>
> Rmsd input:
>
> trajin heat.rmsd
> trajin equil.rmsd
> reference x.inpcrd
> rms reference out backbone.rmsd .CA,C,N
>
>
> [image: Inline image 1]
>
>
> If I remove equil.mdcrd an donly do trajin heat.mdrcd, rmsd is alot lower
> but I can see it is increasing towards the end. So I think it is
> fluctuating more during equilibrium stage. Following is the one only with
> heat.mdcrd
>
>
> [image: Inline image 2]
>
>
>
>
> Thank you all in advance for your time!!!
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Mar 11 2016 - 18:00:08 PST
