[gmx-users] Gromacs 5 umbrella sampling

Justin Lemkul jalemkul at vt.edu
Tue Apr 8 03:17:15 CEST 2014



On 4/4/14, 10:48 AM, Andres Ortega Guerrero wrote:
> Thank you Justin,
>
> I will appreciate the help with this doubt, of the reference group, that is not 0 , is 26 in the index file,
> or the appropiate definition of the reference group thanks
>

I want to revisit this because I am now working through it myself.  What was not 
immediately clear is that these indices do not, in fact, have anything to do 
with the order of the index file.  Instead, the indices correspond to the groups 
used in pulling.

Since the pull code is now more general, such that multiple reaction coordinates 
can now be defined (and thus more than one reference group), one must specify 
the number of reaction coordinates (pull-ncoords), how many groups total are 
involved in any sort of reaction coordinate (pull-ngroups), and then identify 
which of those groups are participating in each of the reaction coordinates.

The simplest case is to use the biasing potential along one reaction coordinate 
with two groups.  Therefore:

pull-ngroups = 2
pull-ncoords = 1
pull-coord1-groups = 1 2
pull-group1-name = refgroup
pull-group2-name = pullgroup

I have that much working at this point.  One could, in theory, then extend the 
situation to have two reaction coordinates between 3 or 4 groups, i.e.

pull-ngroups = 3
pull-ncoords = 2
pull-coord1-groups = 1 2
pull-coord2-groups = 1 3
pull-group1-name = refgroup
pull-group2-name = pullgroup1
pull-group3-name = pullgroup2

I assume that will work (famous last words, but I have no time or need to test 
that setup at the moment).  What it is intended to do is apply a biasing 
potential between groups 1 and 2, and 1 and 3 simultaneously.

-Justin

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

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

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 601
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

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

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


More information about the gromacs.org_gmx-users mailing list