[gmx-users] Analysis of protein-ligand interactions

Justin Lemkul jalemkul at vt.edu
Sun Jul 29 13:15:29 CEST 2012

On 7/29/12 3:30 AM, James Starlight wrote:
> Dear Gromacs Users!
> I want to analyze protein-ligand polar interactions by means of g_hbond utility
> In particular I want to obtain some map wich include information
> aabout dynamics of the h bonds occurence between defined polar groups
> of the protein ( active site) and ligand polar atoms. For that I've
> defined side chains of some amino acinds as the separate group and
> used g_hbond on that group as well as on the ligand. As the result
> I've obtain correct numbers of donor and acceptors as well as average
> number of the h bonds along trajectory. But the map which I need ( I
> suppose that this is the hbmap.xpm file) represent as the long long
> graph with some points ( I suppose that its correspons to the
> different h bonds occured along that trajectory) but without any
> legend on that graph (e.g specified residues of the active site and
> corresonded atoms of the ligand). Is it possible to obtain such
> diagram ?

The key to the figure is in the hbond.ndx file provided by the -hbn argument.

> 2) From Justin's tutorial I've found that the reasonable way to obtain
> information about dominant force mediated protein-ligand interaction
> in particular case is the ussage of G_energy. How from that
> measurements I could conclude which interaction ( e.g vdw or h-bonds)
> is most crussial for the ligand binding ( affinity) ?

Short-range nonbonded interactions (which is what is in the .edr file with 
energygrps) may give you some indication, but this term is often very broadly 
applied when it should not be.  If you want a real analysis of binding energy, 
very careful free energy calculations are necessary.



Justin A. Lemkul, Ph.D.
Research Scientist
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080


More information about the gromacs.org_gmx-users mailing list