[gmx-users] g_mindist , trajectory only CA atoms, to small values?

Justin A. Lemkul jalemkul at vt.edu
Tue Dec 1 23:45:58 CET 2009



t.peter.Mueller at gmx.net wrote:
> Dear Shay, dear Gromacians
> 
>> For purposes of debugging this bug you can also use the flag of g_mindist
>> that gives out atom pairs and see which atom pairs is it that gromacs
>> thinks are less the 1 angstrom.
> I used the  "-o  atm-pair.out" option to get the atom pairs with the minimal
> distance. For the all atom trajectory the correct atom pairs are selected.
> But for the CA trajectory the wrong atom pairs are selected. Instead of the
> pairs with the shortest distance a much further pair is selected. Although

Can you post some of this output?

> the output pair from "atm-pair.out" has a distance of more then 100 Angstrom
> the value in  the file "mindist.xvg" is smaller then one Angstrom.
> 

Is this for the CA-only trajectory?

> I think the "g_mindist" command has a bug. In the trajectory each residue is
> only represented by a single CA atom (to reduce storage footprint and
> increase speed), can it be that g_mindist requires that the atoms have to
> "connected" (no broken bonds)? But for calculating the minimal distance of
> two groups this should really not be required.
> 

Before we jump to the bug conclusion, can you post the command line you're 
using, groups you're selecting (are they custom? simply the default "C-alpha"?) 
and relevant snippets from the output and/or coordinate files that demonstrate a 
bit better what's going on.  If, based on this information, things look weird, 
then file a bugzilla with files that reproduce the problem.  I would disagree 
that g_mindist requires bond information, since a .tpr file is not strictly 
required for its function.

-Justin

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================



More information about the gromacs.org_gmx-users mailing list