[gmx-users] g_hbond

[Erik Marklund]

On Mon, 2006-12-18 at 22:57 +0200, OZGE ENGIN wrote:
> Hi, 
> 
> I use the 3.3.1 version of gromacs.
> 
> My system consists of a capped alanine residue embedded in a box of 500 water molecules. When I select the group 1 option ( only capped alanine residue) there is no error; however, an error occurs due to selection of the whole system (group 0 option). I want to investigate the hydrogen bonds not only formed within the protein but also between the protein molecule (capped alanine residue) and water molecules. 
> 
> Moreover, i could not understand the memory problem. Could you explain it in detail,please?

One thing is that the existence matrix for all the hydrogen bonds is
HUGE if you choose large groups for the analysis. Let's say you have
5000 water molecules and a protein having 100 donors/acceptors, adding
up to a total of 5100 donors/acceptors, and that the number of frames in
your trajectory is 10000. That gives you a maximum size of
5100x5100x10000 = 260100000000 elements. That's a lot of memory.
However, the matrix will never reach that size for two reasons:
1) some potential hydrogen bonds will never form, e.g. because they are
too far apart.
2) not all frames of the existence function is stored in the matrix,
only the ones from the first instance of a particular hydrogen bond to
its last occurrence.

Nevertheless, the calculations illustrate a potential memory problem
arising from having a system too big. This is also why i suggested you
should split the analysis into parts: solvent-solute ...

/Erik

> 
> Thanks in advance 
> 
> Ozge
> 
> -----Original Message-----
> From: Erik Marklund <erikm at xray.bmc.uu.se>
> To: Discussion list for GROMACS users <gmx-users at gromacs.org>
> Date: Mon, 18 Dec 2006 13:48:18 +0100
> Subject: Re: [gmx-users] g_hbond
> 
> On Mon, 2006-12-18 at 13:34 +0200, OZGE ENGIN wrote:
> > Hi,
> >  
> > I have a problem with g_hbond. I want to investigate the H bonds formed not only within the protein but also between water molecules and the protein.In this respect,  I select the group 0 --> 'system' option for both selections, an error message like the following one appears:
> >  
> > Segmentation fault
> >  
> > How can I fix this problem?
> 
> That's hard to say. Which version are you running?
> 
> >  
> > Thanks in advance
> >  
> > Ozge
> > 
> > _______________________________________________
> > 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
> 
-- 
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