# [gmx-users] Re : How GROMACS calculate the energy of hydrogen bond

ABEL Stephane 175950 Stephane.ABEL at cea.fr
Thu May 31 16:51:21 CEST 2012

```Hi,

Here, my 2 cents worth. You can also estimate (roughly !!) the HB energy between two groups (say NH---CO) by using the Kabsch and Sander function  described  in Kabsch, W.; Sander, C.  Biopolymers 1983, 22, 2577−2637).

Quoting:" E = qlq2(1/r(0N) + l/r(CH) - l/r(OH) - l/r(CN))*f with q1 = 0.42e and q 2 = 0.20e, e being the unit electron charge and r(AB) the interatomic distance from A to B. In chemical units, r is in angstroms,
the dimensional factor f = 332, and E is in kcal/mol. A good H bond has about -3 kcal/mol binding energy. We choose a generous cutoff to allow for bifurcated H bonds and errors in coordinates and assign an H bond
between C=O of residue i and N-H of residue j if E is less than the cutoff, i.e., “Hbond(ij)=: [E < -0.5kcal/mole].”

To obtain  HB energy value E, you need only the distance between the donnor and acceptor groups.

HTH

Stephane

------------------------------

Message: 2
Date: Thu, 31 May 2012 15:00:05 +0200
From: "lloyd riggs" <lloyd.riggs at gmx.ch>
Subject: Re: [gmx-users] How GROMACS calculate the energy of hydrogen
bond
To: Discussion list for GROMACS users <gmx-users at gromacs.org>
Message-ID: <20120531130005.302310 at gmx.net>
Content-Type: text/plain; charset="utf-8"

Dear All,

I have no clue what specifically you are trying, but I feal bad for all the physicist and quantum chemist whom have provided the software and continued to develop it.

Scanning in my free time, it seems a large amount of confusion on what people are trying to do stems from differences in what is taught textbook wise for things.

For instance a hydrogen bond to a physicist is an integration over space in 3 dimensions including time and probabilities of occupied spaces (atom position variabilities reflected even more in proteins, ie the necessity of multiple MD runs with different starting conformations), Vs.  an organic chemist whom has cut offs, ie angles between two points and set distances between two atoms which generally reflect the means of calculated chemical energies within a range (say 80-90% which represent means, but usually from raw small molecules as determinants), Vs. Biologist whom have tables which either use a set distance and angle and little account of variability over time (ie a hydrogen bond equals 1.4 kCal/mol reflecting the absolute mean), conformations in amino acids, etc...

I think with gromacs it is very precise, as even the smallest energies between two interacting atoms is taken into account with accuracy reflected by the force fields used, and how they were derived.

Good luck, your going to start seeing more and more a flood of biologist.

Stephan Watkins

-------- Original-Nachricht --------
> Datum: Thu, 31 May 2012 19:54:04 +1000
> Von: Mark Abraham <Mark.Abraham at anu.edu.au>
> An: Discussion list for GROMACS users <gmx-users at gromacs.org>
> Betreff: Re: [gmx-users] How GROMACS calculate the energy of hydrogen bond

> On 31/05/2012 7:46 PM, Acoot Brett wrote:
> > Hi Mark,
> >
> > It is confusing. As you know, for the same hydrogen bond in a protein,
> > the related hydrogen bond angle and bond length can vary within a
> > scope during the whole simulation process, however this small
> > vibration of the hydrogen bond angle and length can lead to
> > significant energy change, and correspondingly the energy of a
> > hydrogen bond in simulation can be varied significantly. In comparison
> > with hydrophobic effect, it would be too much is the energy of the
> > hydrogen bond would be  not calculated  continuously.
>
> It isn't, if the model physics isn't paramtrized to include it
> explicitly - which is the case for all the force fields in GROMACS.
>
> >
> > Could you give some further clarification?
>
> What are trying to do? Measuring "the strength of a hydrogen bond"
> requires you identify a state with and without it and a path between
> them over which you can integrate.
>
> Mark
>
> >
> > Cheers,
> >
> > Acoot
> >
> > ------------------------------------------------------------------------
> > *From:* Mark Abraham <Mark.Abraham at anu.edu.au>
> > *To:* Discussion list for GROMACS users <gmx-users at gromacs.org>
> > *Sent:* Thursday, 31 May 2012 4:48 PM
> > *Subject:* Re: [gmx-users] How GROMACS calculate the energy of
> > hydrogen bond
> >
> > On 31/05/2012 4:42 PM, Acoot Brett wrote:
> >> Dear All,
> >> The value of the energy of the hydrogen bond has relation with
> >> distance and angle of the hydrogen bond related atoms. As for in the
> >> simulation process, the distance and angle of the hydrogen bond
> >> related atoms may change continuously. Will you please let me know
> >> based on which formula GROMACS calculated the value of the energy of
> >> the hydrogen bonds?
> >
> > There is no such formula used in MD force fields implemented in
> > GROMACS. The only non-bonded interactions are the ones you already
> > know about: electrostatics and VDW.
> > Observables like hydrogen bonds and the hydrophobic effect arise from
> > them.
> >
> > Mark
> >
> > --
> > gmx-users mailing list gmx-users at gromacs.org
> > <mailto:gmx-users at gromacs.org>
> > http://lists.gromacs.org/mailman/listinfo/gmx-users
> > Please search the archive at
> > http://www.gromacs.org/Support/Mailing_Lists/Search before posting!
> > Please don't post (un)subscribe requests to the list. Use the
> > www interface or send it to gmx-users-request at gromacs.org
> > <mailto:gmx-users-request at gromacs.org>.
> > Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> >
> >
> >
>
```