[gmx-users] Implicit Solvent seems to collapse molecule to origin

[Justin A. Lemkul]

Andrew Mauro wrote:
> Hi,
> While running a simulation of capped alanine in an implicit solvent the 
> molecule appears to collapse on itself such that the atoms have 
> coordinates of 0,0,0 or some even have NaN.  Has anyone else had this 
> problem or any suggestions?

Sounds like http://www.gromacs.org/Documentation/Terminology/Blowing_Up.

> The simulation was run using GROMACS 4.5 with the amber96ff with 
> brownian dynamics.
> Parameters
> – Langevin dynamics = bd
> – Bd_fric= 95 [amu ps^-1]
> – Implicit_solvent= GBSA
> – gb_altorithm= OBC
> – gb_epsilon_solvent = 80 (dielectric constant of H20)
> – Sa_surface_tension = 2.092 [kJ mol^-1 nm^-2 ]
> – Thermostat= vscale
> 

A complete .mdp file would be vastly more useful than this parsed version. 
There are numerous factors that will influence the stability of your system, 
many of which are omitted here.

You can also try upgrading to version 4.5.3, as there have been several bug 
fixes and improvements to the implicit solvent code (among others).

-Justin

> Thanks for any help.
> 
> -- 
> Andrew Mauro
> 

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

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

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