[gmx-users] MD exit condition

[Justin Lemkul]

On 7/19/15 9:16 PM, Eric Smoll wrote:
> Hello Gromacs users,
>
> In response to a question on setting a group-group distance as an MD exit
> condition, Mark Abraham suggested that it might be possible to use a
> table-specified bond that would terminate a simulation by generating an
> infinite force and crashing mdrun.
>
> I am interested in revisiting this idea with a slightly more complex exit
> condition. How would I define a simulation that will crash when the
> z-component of a fictious bond between two groups exceeds a certain value
> in a specific cartesian direction (e.g., the z-component)? Is this
> impossible in GROMACS?
>

Not without code modifications.  Nothing in the mdrun code checks distances or 
components of any vector to determine whether or not to stop.  That would be a 
drag on performance for something that wouldn't really get used that often. 
Shouldn't be hard to add.

-Justin

> Best,
> Eric
>
> On Thu, Aug 28, 2014 at 6:58 AM, Mark Abraham <mark.j.abraham at gmail.com>
> wrote:
>
>> On Thu, Aug 28, 2014 at 3:32 PM, Eric Smoll <ericsmoll at gmail.com> wrote:
>>
>>> Hello Gromacs users,
>>>
>>> I am using Gromacs 4.6.5
>>>
>>> Is there a way to set an exit condition during a simulation? I am
>>> interested in terminating dynamics when two groups are separated by a
>>> certain distance.
>>>
>>
>> No. People want this, but it's never been a priority for someone with the
>> knowledge to implement it. Creative use of a bond with a "malware" user
>> table could produce infinite forces that will get mdrun to blow up shortly
>> after the condition is reached, but I can't honestly recommend that! :-D
>>
>> One reasonable approach might be to use the pull code to achieve the
>> distance in a controlled way.
>>
>> If it is not possible to set an exit condition, I suspect I can post
>>> process the trajectory with trjconv using the -exec flag and/or the
>> -trunc
>>> flag:
>>>
>>> 1.) The manual states that the -exec flag will "Execute a command for
>> every
>>> output frame with the frame number as argument." What kinds of commands I
>>> can pass to this flag?
>>>
>>
>> Intended for use with -sep or -split. Command gets passed the number of the
>> file most recently written, and then has to go do its own file parsing. I
>> would rather use g_dist or such.
>>
>> 2.) If the "-exec" is not useful, I plan to analyse the the output of
>>> g_dist to find the the time "t" when two groups exceed my distance
>>> criterion and run "trjconv -trunc t" to crop the trajectory
>> appropriately.
>>> Is there a better way?
>>>
>>
>> You could probably use NAMD :-)
>>
>> Mark
>>
>>
>>> Best,
>>> Eric
>>> --
>>> Gromacs Users mailing list
>>>
>>> * Please search the archive at
>>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>>> posting!
>>>
>>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>>
>>> * For (un)subscribe requests visit
>>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>>> send a mail to gmx-users-request at gromacs.org.
>>>
>> --
>> Gromacs Users mailing list
>>
>> * Please search the archive at
>> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
>> posting!
>>
>> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>>
>> * For (un)subscribe requests visit
>> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
>> send a mail to gmx-users-request at gromacs.org.
>>

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

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

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