[gmx-users] g_hbond

David van der Spoel spoel at xray.bmc.uu.se
Thu Aug 10 20:12:49 CEST 2006


Chinmay Das wrote:
> Dear David,
>  ok. Now I understand HDA criterion. Can you give me a reference
> where I can find the differences between HDA criteria and OHD criteria
> and why one of them is preferable than the other?

We implemented this in order to reproduce a number of publications. It 
is difficult to say that one is to be preferred over the other.  If 
something an energy-based criterium could be preferred, however 
classical mechanics is not good enough to quantitatively describe the 
details of hydrogen bonding anyway. Some interesting refs are:

@Article{Starr2000a,
   author = 	 {F. W. Starr and J. K. Nielsen and H. Eugene Stanley},
   title = 	 {Hydrogen-bond dynamics for the extended simple point 
charge model of water},
   journal = 	 {Phys. Rev. E},
   year = 	 2000,
   volume =	 62,
   pages =	 {579-587}
}

@Article{Modig2003a,
   author = 	 {K. Modig and B. G. Pfrommer and B. Halle},
   title = 	 {Temperature-dependent hydrogen bond geometry in liquid water},
   journal = 	 {Phys. Rev. Lett.},
   year = 	 2003,
   volume =	 90,
   pages =	 075502
}

@Article{Spoel2006b,
   author = 	 {D. van der Spoel and P. J. van Maaren and P. Larsson and 
N. Timneanu},
   title = 	 {Thermodynamics of hydrogen bonding in hydrophilic and
hydrophobic media},
   journal =      {J. Phys. Chem. B},
   year =         2006,
   volume =       110,
   pages = 	 {4393-4398}
}


>  vmd certainly uses the later criteria. Changing the angle criteria
> there  44 degrees in vmd picks up the bond that it was missing
> earlier.
> best regards
> Chinmay
> 
> On 8/10/06, David van der Spoel <spoel at xray.bmc.uu.se> wrote:
>> Chinmay Das wrote:
>> > Dear Erik,
>> >  Thanks a lot. with -nomerge, atleast I can see which atoms are 
>> involved.
>> > [10 12 22] triplet, yes, distance is 0.297. But it should not go
>> > through the angle criteria.
>> > R(10,12) = {0.088, 0.013, - 0.046}, |R(10,12)| = 0.10
>> > R(12,22) = {0.209, -0.018, 0.052}, |R(12,22)|=0.216
>> > R(10,12).R(12,22)=0.015766
>> > cos(angle)=R1.R2/(|R1| |R2|) = 0.7284
>> > ie, the angle 43.2 degrees.
>> > How did you get the angle 29.89 degrees?
>>
>> you're computing a different angle. g_hbond uses HDA, or in your
>> numbering 12-10-22, that is you have to compute vectors 10-12 and 10-22
>>
>>
>> > best regards
>> > Chinmay
>> >
>> > On 8/10/06, Erik Marklund <erikm at xray.bmc.uu.se> wrote:
>> >
>> >> It was a hard nut to crack, but I think I've done it after some 
>> sifting
>> >> the source code:
>> >>
>> >> Since the -merge option is switched on by default, the hydrogen 
>> indexes
>> >> are more or less irrelevant for some parts of the code. What happens
>> >> here is that g_hbond finds the hbond [10 12 22] (distance: 0.2971
>> >> nm, angle: 29.8925), but at a later point changes it into [10 11 22]
>> >> during the merging, just like [22 24 16] turns into [22 23 16]. Note
>> >> that the latter is apparent from the difference between what you find
>> >> and the .ndx-file given below. Try running g_hbond -nomerge and you 
>> will
>> >> see a slightly different ndx-file.
>> >>
>> >> One could debate whether this is really a bug or not, and the merging
>> >> should in principle behave like this. I guess that an alternative 
>> is to
>> >> completely omit the hydrogens when building the ndx-file using 
>> -merge to
>> >> avoid confusion.
>> >>
>> >> A question arises: Why do VMD and yourself both find only 3 hbonds 
>> when
>> >> there are four of them?
>> >>
>> >> /Erik
>> >>
>> >> On Wed, 2006-08-09 at 15:17 +0100, Chinmay Das wrote:
>> >> > Hi,
>> >> >   I am having trouble understanding the criteria for hydrogen 
>> bond and
>> >> > how it is implemented in g_hbond. Defining hydrogen bond for 
>> O1--H..O2
>> >> > by requiring distance(O1, O2) < 0.35 nm and angle(O1-H, H-O2) < 30
>> >> > degrees, my own code consistently give lower number of hydrogen 
>> bonds
>> >> > than g_hbond. But my results match if I count the hydrogen bonds
>> >> > plotted with vmd using the same criteria. (To test this I created a
>> >> > small system of water molecules at the center of a large box - so
>> >> > periodic boundary condition is not the reason.)
>> >> > Can some one help?
>> >> >
>> >> > Some details: I placed randomly oriented 8 SPC water on the vertices
>> >> > of a cube with sides of size 0.3, starting at {x,y,z}={1.5,1.5,1.5}
>> >> > and set the box length to be {4, 4, 4}. A very short NVT run at 30K
>> >> > orients the water but doesn't move them much. vmd shows that the
>> >> > configuration has 3 hydrogen bonds and my calculation also picks up
>> >> > the same triplets and identifies as hydrogen bond. However, g_hbond
>> >> > -hbn picks up 4. The hbonds_SOL entry in hbond.ndx have two of the
>> >> > correct entries and two entries which seem wrong.
>> >> >
>> >> > hbond.ndx entry :
>> >> > [ hbonds_SOL ]
>> >> >       1      2     13
>> >> >      10     11     22
>> >> >      16     17     13
>> >> >      22     23     16
>> >> > What I find :
>> >> > 1-2 -> 13
>> >> > 16-17 -> 13
>> >> > 22-24 -> 16
>> >> >
>> >> > And the configuration file :
>> >> > Pure Water
>> >> >    24
>> >> >     1SOL     OW    1   1.494   1.496   1.495 -0.2407 -0.3859 -0.4517
>> >> >     1SOL    HW1    2   1.593   1.511   1.498 -0.6476  3.2879  3.3372
>> >> >     1SOL    HW2    3   1.470   1.416   1.551  0.4456  0.5879  1.2955
>> >> >     2SOL     OW    4   1.506   1.503   1.801  0.3278  0.0764  0.1469
>> >> >     2SOL    HW1    5   1.586   1.482   1.857  0.3064  1.6146  0.8017
>> >> >     2SOL    HW2    6   1.435   1.434   1.817  0.1439  0.6902  2.4044
>> >> >     3SOL     OW    7   1.499   1.810   1.507 -0.1617  0.6516  0.4652
>> >> >     3SOL    HW1    8   1.446   1.767   1.579 -0.1540 -3.2274 -1.5277
>> >> >     3SOL    HW2    9   1.439   1.833   1.430  0.5590 -2.0347 -1.0463
>> >> >     4SOL     OW   10   1.499   1.802   1.796 -0.0922  0.1504 -0.2160
>> >> >     4SOL    HW1   11   1.501   1.717   1.849 -2.8903 -1.0982 -1.9099
>> >> >     4SOL    HW2   12   1.587   1.815   1.750  1.3885 -0.2196  2.3356
>> >> >     5SOL     OW   13   1.798   1.491   1.501 -0.1855 -0.6306 -0.0502
>> >> >     5SOL    HW1   14   1.856   1.410   1.508  2.6216  1.0980 -1.5287
>> >> >     5SOL    HW2   15   1.813   1.535   1.412  0.8590  3.2110  1.8097
>> >> >     6SOL     OW   16   1.804   1.499   1.803  0.1303 -0.0398  0.1705
>> >> >     6SOL    HW1   17   1.771   1.492   1.709  2.6087  3.3512 -1.1639
>> >> >     6SOL    HW2   18   1.865   1.423   1.824 -0.0578 -0.5609 -1.1170
>> >> >     7SOL     OW   19   1.806   1.799   1.496  0.4549 -0.1262 -0.0397
>> >> >     7SOL    HW1   20   1.762   1.879   1.537 -3.4166  1.6216 -6.2709
>> >> >     7SOL    HW2   21   1.754   1.717   1.521 -3.2644  1.1647 -2.6085
>> >> >     8SOL     OW   22   1.796   1.797   1.802 -0.2321 -0.1817  0.1797
>> >> >     8SOL    HW1   23   1.830   1.847   1.722  3.7237 -1.1847  0.9954
>> >> >     8SOL    HW2   24   1.808   1.699   1.788 -2.2397 -0.5892  0.9430
>> >> >    4.00000   4.00000   4.00000
>> >> >
>> >> > Thanks in advance.
>> >> > Chinmay
>> >> > _______________________________________________
>> >> > gmx-users mailing list    gmx-users at gromacs.org
>> >> > http://www.gromacs.org/mailman/listinfo/gmx-users
>> >> > Please don't post (un)subscribe requests to the list. Use the
>> >> > www interface or send it to gmx-users-request at gromacs.org.
>> >> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>> >> --
>> >> Erik Marklund, PhD Student, Molecular Biopcysics group,
>> >> Dept. of Cell and Molecular Biology, Uppsala University.
>> >> Husargatan 3, Box 596,          75124 Uppsala, Sweden
>> >> phone: +46 18 471 4537          fax: +46 18 511 755
>> >> erikm at xray.bmc.uu.se
>> >>
>> >> _______________________________________________
>> >> gmx-users mailing list    gmx-users at gromacs.org
>> >> http://www.gromacs.org/mailman/listinfo/gmx-users
>> >> Please don't post (un)subscribe requests to the list. Use the
>> >> www interface or send it to gmx-users-request at gromacs.org.
>> >> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>> >>
>> > _______________________________________________
>> > gmx-users mailing list    gmx-users at gromacs.org
>> > http://www.gromacs.org/mailman/listinfo/gmx-users
>> > Please don't post (un)subscribe requests to the list. Use the www
>> > interface or send it to gmx-users-request at gromacs.org.
>> > Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
>>
>> -- 
>> David.
>> ________________________________________________________________________
>> David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
>> Dept. of Cell and Molecular Biology, Uppsala University.
>> Husargatan 3, Box 596,          75124 Uppsala, Sweden
>> phone:  46 18 471 4205          fax: 46 18 511 755
>> spoel at xray.bmc.uu.se    spoel at gromacs.org   http://folding.bmc.uu.se
>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>> _______________________________________________
>> gmx-users mailing list    gmx-users at gromacs.org
>> http://www.gromacs.org/mailman/listinfo/gmx-users
>> Please don't post (un)subscribe requests to the list. Use the
>> www interface or send it to gmx-users-request at gromacs.org.
>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
> _______________________________________________
> gmx-users mailing list    gmx-users at gromacs.org
> http://www.gromacs.org/mailman/listinfo/gmx-users
> Please don't post (un)subscribe requests to the list. Use the www 
> interface or send it to gmx-users-request at gromacs.org.
> Can't post? Read http://www.gromacs.org/mailing_lists/users.php


-- 
David.
________________________________________________________________________
David van der Spoel, PhD, Assoc. Prof., Molecular Biophysics group,
Dept. of Cell and Molecular Biology, Uppsala University.
Husargatan 3, Box 596,  	75124 Uppsala, Sweden
phone:	46 18 471 4205		fax: 46 18 511 755
spoel at xray.bmc.uu.se	spoel at gromacs.org   http://folding.bmc.uu.se
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++



More information about the gromacs.org_gmx-users mailing list