[gmx-users] Re: Bonds break while Minimising using distance restraints

plmallip at mail.uh.edu plmallip at mail.uh.edu
Thu Sep 4 01:30:53 CEST 2008


Hi,


The atoms are separated by 12 A (1.2
> nm). The atoms I restrained are 1703 & 1712 (6151 & 6160 atoms are also
> from the same residues). These are the residues that looks as if they
> are no bonds between them. ( residues alone, peptide bond alone and the
> rest of the protein).

How is it that atoms that are 900 indices apart are in the same residues.  I
don't understand.  Do you mean that all four of these atoms are in the same
residue?  Then they definitely aren't amino acids!  Are these two separate
residues bridged by a distance restraint, that you are considering one residue?
  Is there some bond defined in the topology between *any* of these atoms?

My protein has 5 chains. 1703 & 1712 are the numbers from chainC.itp file. This includes the distance restraints -

#ifdef DDISRES
[ distance_restraints ]
;ai   aj type index type' low   up1  up2  fac
1703 1712 1    0     1    1.15  1.20 1.25 1.0
#endif

6151 & 6160 atoms are the numbers from .gro file reported by the .job file. 6151 is N of residue 584 and 1703 is CA of the same redsidue. Residue 584 & 585 are placed have 14 residues missing between them. 6160 is N of residue 585 &  1712 is the corresponding CA.

>Then, are you correctly applying "define = -DDDISRES" in your .mdp as I
>suggested previously?

I did correctly use the -DDDISRES, as suggested by you, which is as follows:

title               =  ${MOL}
cpp                 =  /lib/cpp
define              = -DDDISRES


Thanks & regards,
Latha.


----- Original Message -----
From: "Justin A. Lemkul" <jalemkul at vt.edu>
Date: Wednesday, September 3, 2008 5:36 pm
Subject: Re: Bonds break while Minimising using distance restraints
To: plmallip at mail.uh.edu, Gromacs Users' List <gmx-users at gromacs.org>

> 
> 
> plmallip at mail.uh.edu wrote:
> > Hi Justin,
> > 
> >       Thanks for your prompt 
> suggestions.  I hope I didn't cause any 
> > sort of inconvenience by E-mailing you directly. 
> > 
> 
> I say this constantly - it is always better to keep the 
> discussion on the list 
> so that, if I don't have the complete right answer (which 
> happens often!) 
> someone else can weigh in and help out.  It is also helpful 
> to complete 
> discussions in the archives.  There's nothing more 
> frustrating than finding your 
> exact problem in the archives, only to find out that the thread 
> dies off without 
> a solution :)
> 
> >  >I thought you were applying a 12-nm distance restrain 
> to atoms in the 
> > 7000's, so
> >  >I'm confused as to why atoms615 1 and 6160 should be 12 
> nm apart 
> > (assuming, of
> >  >course, that you meant to say nm instead of A :)
> > 
> >      I am sorry for the confusion. 
> The atoms are separated by 12 A (1.2 
> > nm). The atoms I restrained are 1703 & 1712 (6151 & 6160 atoms 
> are also 
> > from the same residues). These are the residues that looks as 
> if they 
> > are no bonds between them. ( residues alone, peptide bond 
> alone and the 
> > rest of the protein).
> 
> How is it that atoms that are 900 indices apart are in the same 
> residues.  I 
> don't understand.  Do you mean that all four of these atoms 
> are in the same 
> residue?  Then they definitely aren't amino acids!  
> Are these two separate 
> residues bridged by a distance restraint, that you are 
> considering one residue? 
>   Is there some bond defined in the topology between *any* 
> of these atoms?
> 
> > 
> > While doing grompp I get the following "removed 1 distance 
> restraints". 
> > Does this mean, the distance restraint isn't imposed at all?
> > 
> 
> That would be a pretty clear indication to me that distance 
> restraints are not 
> being applied.
> 
> >  I corrected the distance restraints I used as
> > 
> > #ifdef DDISRES
> > [distance_restraints]
> > ;ai   aj type index type' low up1   
> up2  fac
> > 1703 1712 1    0     
> 1    1.15  1.20 1.25 1.0
> > #endif
> > 
> 
> Then, are you correctly applying "define = -DDDISRES" in your 
> .mdp as I 
> suggested previously?
> 
> http://www.gromacs.org/pipermail/gmx-users/2008-September/036136.html
> 
> -Justin
> 
> > I endup with the same error. Is there any possible way to get 
> over this 
> > problem?
> > 
> > I appreciate your help in this regard.
> > 
> > wamr regards,
> > Latha.
> > 
> > 
> >  >      There is no chance of 
> steric clashes between 6151 & 6160. They are
> >  > seperated by 12 A. ("Warning: 1-4 interaction between 
> 6151 and 6160 at
> >  > distance 1.387 which is larger than the 1-4 table size 
> 1.000 nm. These
> >  > are ignored for the rest of the simulation. This 
> usually means your
> >  > system is exploding") I cannot build the missing 
> residues without
> >  > knowing the secondary structure. I am already running 
> one simulation
> >  > with built residues. But, my ultimate goal is to run 
> dynamics with
> >  > distance restraints.
> > 
> > Are you sure you are identifying the correct atoms?  A 1-
> 4 interaction 
> > involves
> > atoms that are very close together (in fact, three bonds 
> away).  Are you 
> > saying
> > that they are separated by 12 A (1.2 nm) or by 12 nm?  If 
> they are 
> > starting 12 A
> > (1.2 nm) away, then the error message still makes sense.  
> If they are 12 nm
> > away, then I suspect that some sort of bond or restraint is 
> being applied
> > improperly.
> > 
> >             
> > 
> > I thought you were applying a 12-nm distance restrain to atoms 
> in the 
> > 7000's, so
> > I'm confused as to why atoms615 1 and 6160 should be 12 nm 
> apart 
> > (assuming, of
> > course, that you meant to say nm instead of A :)
> > 
> > -Justin
> > 
> >  >
> >  >        My sytem is 
> an pentamer. Here is the toplogy file
> >  >
> >  > ; Include forcefield parameters
> >  > #include "ffG43a1.itp"
> >  >
> >  > ; Include chain topologies
> >  > #include "chnrc_A.itp"
> >  > #include "chnrc_B.itp"
> >  > #include "chnrc_C.itp"
> >  > #include "chnrc_D.itp"
> >  > #include "chnrc_E.itp"
> >  >
> >  > ; Include water topology
> >  > #include "spce.itp"
> >  >
> >  > #ifdef POSRES_WATER
> >  > ; Position restraint for each water oxygen
> >  > [ position_restraints ]
> >  > ;  i funct       
> fcx        fcy        fcz
> >  >    1    
> 1       1000       1000       1000
> >  > #endif
> >  >
> >  > ; Include generic topology for ions
> >  > #include "ions.itp"
> >  >
> >  > [ system ]
> >  > ; Name
> >  > Protein in water
> >  >
> >  > [ molecules ]
> >  > ; Compound        #mols
> >  > 
> Protein_A           1
> >  > 
> Protein_B           1
> >  > 
> Protein_C           1
> >  > 
> Protein_D           1
> >  > 
> Protein_E           1
> >  > 
> SOL              55419
> >  > 
> NA+              54
> >  >
> >  >
> >  > Thanks & regards,
> >  > Latha.
> >  >
> >  >  > > plmallip at mail.uh.edu wrote:
> >  >  > >
> >  >  > > Hi Justin, thanks for your response. You are 
> right. I get the
> >  >  > following
> >  >  > > message in the .job file
> >  >  > >
> >  >  > > "Warning: 1-4 interaction between 6151 and 
> 6160 at distance
> >  >  > 1.387 which
> >  >  > > is larger than the 1-4 table size 1.000 nm
> >  >  > > These are ignored for the rest of the simulation
> >  >  > > This usually means your system is exploding"
> >  >  > >
> >  >  > > The atoms I restrained are 1703 & 1712 (6151 
> & 6160 atoms are
> >  >  > also from
> >  >  > > the same residues). The distance between 
> them is 12 A (there
> >  >  > are 14
> >  >  > > missing residues in between). My aim is to 
> use distance
> >  >  > restraints,
> >  >  > > without building missing residues in 
> between. Is there any way
> >  >  > I can
> >  >  > > overcome this warning why minimizing the system?
> >  >  >
> >  >  > It's very hard to say, because none of us 
> knows what's in your
> >  >  > system, topology,
> >  >  > or how you built things :)
> >  >  >
> >  >  > You have some type of nasty steric clash 
> that's driving atoms
> >  >  > 6151 and 6160
> >  >  > apart.  Visualize the trajectory (.trr) 
> to see if you can
> >  >  > identify where things
> >  >  > start to break down.  I don't know if the 
> distance>  >  > restraint has anything to do
> >  >  > with the problem or not.
> >  >  >
> >  >  > Is there a problem with building missing 
> residues?  That
> >  >  > might make life quite a
> >  >  > bit easier in the long run.
> >  >  >
> >  >  > -Justin
> >  >  >
> >  >  > >
> >  >  > > Thanks & regards,
> >  >  > > Latha.
> >  >  > >
> >  >  > >
> >  >  > >  > Dear colleagues,
> >  >  > >  >
> >  >  > >  > I need to use simple distance 
> restraints of 12.5 A
> >  >  > between two CA atoms
> >  >  > >  > of two residues. I am using the 
> following lines in the
> >  >  > .itp file
> >  >  > >  >
> >  >  > >  > #ifdef DDISRES
> >  >  > >  > [distance_restraints]
> >  >  > >  > ;ai   aj type index type' low
> >  >  > up1  up2  fac
> >  >  > >  > 1703 1712 1
> >  >  > 0     1    
> 11.5  12.0
> >  >  > 12.5 1.0
> >  >  > >  >
> >  >  > >  > The first feww line of .mdp file for 
> minimisation of
> >  >  > protein alone is
> >  >  > >  >
> >  >  > >  > ; Preprocessing
> >  >  > >  > ;
> >  >  > >  >
> >  >  > 
> title               =  ${MOL}
> >  >  > >  >
> >  >  > 
> cpp                 =  /lib/cpp    ;Preprocessor
> >  >  > >  >
> >  >  > 
> define              = -DDISRES  ;For cg, and also steep
> >  >  > >  >
> >  >  > >
> >  >  > > You are not actually applying your distance 
> restraint.>  >  > If you have "#ifdef
> >  >  > > DDISRES," then you would have to "define = -
> DDDISRES" in the
> >  >  > .mdp file.
> >  >  > > What
> >  >  > > you probably meant to define was "#ifdef 
> DISRES" in the topology.
> >  >  > >
> >  >  > >  >
> >  >  > After minimisation, the restarined residues & 
> the adjacent
> >  >  > >  > bonds break. This results in 
> fragmnets - residues
> >  >  > alone, peptide bond
> >  >  > >  > alone and the rest of the protein. 
> The distance
> >  >  > restrained residues
> >  >  > >  > seems to try to move towards each 
> other (6.04 A after
> >  >  > minimisation) and
> >  >  > >  > this might have caused 
> fragmentation. I tried to use
> >  >  > various upper and
> >  >  > >  > lower values for bond length so as 
> to increase
> >  >  > flexibility. But, still I
> >  >  > >  > end up in the fragments.
> >  >  > >  >
> >  >  > >
> >  >  > > Bonds don't break in classical MD, this is 
> just an artifact of
> >  >  > > visualization,
> >  >  > > probably from nasty steric clashes within 
> your structure.
> >  >  > >
> >  >  > >
> >  >  > >  
> >         Some
> >  >  > of the suggestions in the archive says VMD 
> doesn't show
> >  >  > >  > bonds if they r above threshold 
> value. When I checked
> >  >  > the distances
> >  >  > >  > between the atoms, one of them is 
> really long CO-CA
> >  >  > bond 3.23 A
> >  >  > >  > (normally its 1.59A). This means the 
> bond is no longer
> >  >  > there.>  >
> >  >  > >
> >  >  > > No, the bond is there, VMD just isn't smart 
> enough to see it
> >  >  > :)  You are
> >  >  > > probably well on your way to an explosion if 
> you try to
> >  >  > constrain bond
> >  >  > > lengths
> >  >  > > with LINCS, however.
> >  >  > >
> >  >  > > -Justin
> 
> -- 
> ========================================
> 
> Justin A. Lemkul
> Graduate Research Assistant
> Department of Biochemistry
> Virginia Tech
> Blacksburg, VA
> jalemkul[at]vt.edu | (540) 231-9080
> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
> 
> ========================================
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20080903/54685c37/attachment.html>


More information about the gromacs.org_gmx-users mailing list