[gmx-users] constraints

[Mark Abraham]

rams rams wrote:
> Dear Gromacs Users,
> 
> I have a problem in running the energy minimization of a protein along 
> with a substrate. In the .mdp file, if I use constraints =  all-bonds, 
> the system complains like the convergence is reached to system accuracy 
> not to the one I mentioned.

I don't understand what you mean. It's usually best to copy and paste an 
appropriate small amount of actual input and/or output, rather than 
filter it through two people's understanding of GROMACS and English. :-)

> If I replace the constraints = none, its 
> running well. I would like to know how serious it is to replace 
> all-bonds constraints with none.

If your system has bond lengths that already resemble the equilibrium 
ones, and the system is not really large, and you're just preparing a 
structure for MD, then for EM it really doesn't matter whether you use 
constraints.

Mark