[gmx-users] g_hbond

[Mark Abraham]

On 1/09/2011 6:14 PM, Steven Neumann wrote:
>
>
> On Wed, Aug 31, 2011 at 5:54 PM, Justin A. Lemkul <jalemkul at vt.edu 
> <mailto:jalemkul at vt.edu>> wrote:
>
>
>
>     Steven Neumann wrote:
>
>         One question.... for Glycine it is easy to assess 3 possible
>         hbonds which can create as hydrogen is only one atom as a side
>         chain.
>         How about other amino acids and their maximum hbonds they can
>         create?
>
>
>     Any OH or NH group is a donor, any lone pair is an acceptor
>     (though obviously not modeled explicitly in MD).  The ability of
>     MD force fields to agree with reality in this respect is
>     debatable, but should come close.
>
>     -Justin
>
>     Thank you Justin. Can you please clarify me something and tell me
>     whether I am wrong. This is what I obtained from calulating hbonds
>     between 10 ligands and Glycine. Each ligand serve 8 OH group
>     (flavanoid) :
>
> /*Reading file md.tpr, VERSION 4.5.4 (single precision)
> Specify 2 groups to analyze:
> Group     0 (           r_96) has     7 elements*     -/ this is my 
> Glycine - 7 atoms (side chain - Hydrogen)
> /*Group     1 (            LIG) has   510 elements*    -/ these are 10 
> ligands, 51 atoms each, 8 OH group
> /*Select a group: 0 1
> Selected 0: 'r_96'
> Select a group: Selected 1: 'LIG'
> Checking for overlap in atoms between r_96 and LIG
> Calculating hydrogen bonds between r_96 (7 atoms) and LIG (510 atoms)
> Found 81 donors and 112 acceptors */
> *//*
> /81 donors? It should be 80 when I have 10 ligands with 8 Oh group... 
> Am I right?/

You are assuming g_hbond knows what you know - that only your ligand 
hydroxyls can do effective H-bonding. However, g_hbond can only look at 
the atom names of the groups you give it. Algorithms do what you say, 
not what you mean, unfortunately. 81 = 1*1 + 8*10, since NH from glycine 
is a donor.

> /112 acceptors? 7 atoms of my Glycine x 12 possibilities?/

Methylene atoms cannot accept hydrogen bonds! See g_hbond -h. N and O 
from glycine are acceptors. 112 = 2*1 + 11*10

Mark

> *//*
> /*trn version: GMX_trn_file (single precision)
> Reading frame       0 time    0.000
> Will do grid-seach on 11x11x11 grid, rcut=0.35
> Last frame       2000 time 100000.000
> Average number of hbonds per timeframe 0.642 out of 4536 
> possible     - *that is understood = (112x81)/2/
> //
> /So how many possibilities has Glycine in order to create hbond? Shall 
> I choose just a side chain which is hydrogen?/
> //
> /Steven/
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110901/fcf7fa41/attachment.html>