[gmx-users] helices, again

[Nagy, Peter I.]

Dear Gromacs Users!

  I apologize if experts are fed up with my questions, but I am a beginner
and would like to move ahead.

  I found a strange thing when I performed energy minimization. I created a
conf.gro file from the pdb input using pdb2gmx and also obtained topol.top

Then I typed

grompp -f emgas.mdp -c conf.gro -p topol.top

(emgas.mdp file see below). The box vectors (bx, by, bz) in conf.out were 

Bx = 2.29, By = 4.74 Bz = 2.50

As the result of EM, I obtained the confout.gro file. I compared conf.gro with confout.gro. The first column with real numbers (I assumed that they are 
the x-coordinates) changed only by thousands of an nm (changes in the third
decimal). The difference of the third columns (z-coordinates) of conf.gro
and confout.gro was about 2Bz for the firs helix !!!. x and y coordinates 
differed by about 0.001-0.005 nm for the first helix. When I moved to the
second one, then x changed in the two .gro's by up to 0.005 nm, now  
y differed by "By" (it is not a joke with the word by, I meant the difference
was about By = 4.7 nm), and the z difference was reduced to 1Bz.

  Thus it seems to me that the coordinates were modified by 2Bz for
helix1, and by (1By + 1Bz) for helix2. Then I am not surprised that when
I performed an MD and checked the motion in ngmx, my double helix seemed to
have fallen apart. The very likely reason is a coordinate transformation
throughout the energy minimization. The minimization process itself probably works correctly, because the coordinate changes only by up to 0.01 nm, and the rude change is a consistent translation along the y and z axes. But what did cause it?

  For safety, I enclose the emgas.mdp file below:

title               =  cpeptide
cpp                 =  /lib/cpp
constraints         =  none
integrator          =  steep
dt                  =  0.002    ; ps
nsteps              =  1000
nstlist             =  10
ns_type             =  grid
rlist               =  0.9
rcoulomb            =  0.9
rvdw                =  0.9
;
;       Energy minimizing stuff
;
emtol               =  100.0
emstep              =  0.01


Thanks for any help

Peter Nagy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20030715/dc717f6d/attachment.html>