[gmx-users] problem

Justin Lemkul jalemkul at vt.edu
Sun May 6 16:42:57 CEST 2018

On 5/6/18 9:47 AM, Alex wrote:
> Thanks.
> On Sun, May 6, 2018 at 9:37 AM, Justin Lemkul <jalemkul at vt.edu> wrote:
>> On 5/6/18 9:31 AM, Alex wrote:
>>> Hi
>>> On Sat, May 5, 2018 at 1:14 PM, Mark Abraham <mark.j.abraham at gmail.com>
>>> wrote:
>>> Hi,
>>>> I don't know of any forcefield that uses such nonbonded cutoffs, but that
>>>> should probably not of itself lead to instability. It will just be harder
>>>> to convince your reviewers that your simulation models reality.
>>>> I do not understand what you mean, you mean the 1.3 nm for the rcoulomb
>>> and
>>> rvdw is not a good choice?
>> What prompted you to choose that value? Cutoffs are part of the force
>> field; if you set values that are not in accordance with the force field's
>> underlying parametrization convention, you can invalidate the physical
>> model. rcoulomb is somewhat flexible when using PME, but in general, use
>> what is prescribed unless you have demonstrated that whatever you're doing
>> is superior.
> I see, I always imagined the higher the cut-off value the better, and did

That's a common mistake but certainly untrue.

> not pay attention that it need also to fit with the used FF. Here I am
> using the gromos54a7 FF. I am going to reduce it to around 0.9. Although I

rvdw needs to be 1.4 for GROMOS force fields.

> do not think it can help with the LINC warning and crashing. So, do you
> know what else might cause the problem here in spite of several round of
> successful equilibration.

Hard to say based on such limited information, but there's no reason to 
assume that "previous things worked so the next things will." General 
tips for troubleshooting are found at 



Justin A. Lemkul, Ph.D.
Assistant Professor
Virginia Tech Department of Biochemistry

303 Engel Hall
340 West Campus Dr.
Blacksburg, VA 24061

jalemkul at vt.edu | (540) 231-3129


More information about the gromacs.org_gmx-users mailing list