[gmx-users] Inconsistent shifts Problem when using g_rdf

[Berk Hess]

> Date: Fri, 16 Jan 2009 09:06:43 +0100
> From: spoel at xray.bmc.uu.se
> To: gmx-users at gromacs.org
> Subject: Re: [gmx-users] Inconsistent shifts Problem when using g_rdf
> 
> Lin, Pohan wrote:
> > I am running NPT simulations by using Parrinello-Rahman pressure 
> > coupling on Gromacs 3.3.3.  My system contains epoxy and cross-linker 
> > molecules.  By modifying the topology file (add bond, angle, torsion 
> > terms in topology), all the molecules are connected together to be an 
> > huge polymeric molecule.  The MD jobs were running well before and after 
> > polymerization.  However, while analyzing the data by using g_rdf,  it 
> > keeps showing the message “There were XXX inconsistent shifts.  Check 
> > your topology” for the structures after polymerization and the g(r) are 
> > not reasonable.  Does anyone have any idea about this issue?   I am also 
> > curious about how g_rdf deal with “inconsistent shifts.”  Does g_rdf 
> > ignore those data or do something else?
> 
> Probably, yes. I think it will go fine in version 4.0 though. Please 
> wait until 4.0.3 before upgrading.

Many analysis tools will give these inconsistent shift errors when you
have periodic molecules in the system.
This has not changed in version 4.0.

g_rdf only uses requires the shifting for removing pairs that are excluded
through bonded interactions.
If you do not encounter such pairs in your analysis, or you don't care about
the peaks they will cause at short distance, you can supply only an
index file to g_rdf and no tpr file. In that case the shift code will not be used.

Berk


_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20090116/89da8193/attachment.html>