Hi Martine,
I have done some modeling using NMR restraints. The following files show an
example that I modeled successfully using MODELLER.
The following is the *.top file:
# Homology modelling by the MODELLER TOP routine model
INCLUDE # Include predefined TOP routines
SET TOPLIB = '${LIB}/top_allh.lib' # topology library
SET TOPOLOGY_MODEL = 1 # corresponding topology model;
SET PARLIB = '${LIB}/par.lib' # parameters library
SET HYDROGEN_IO = 'ON' # include hydrogen atoms
SET HETATM_IO = 'ON' # include HETATM entries
SET ALNFILE = '2ptl.ali' # alignment filename
SET KNOWNS = '2ptl' # template name
SET SEQUENCE = '2igh' # target sequence
CALL ROUTINE = 'model' # do homology modelling
# Redefine the rd_restraints routine:
SUBROUTINE ROUTINE = 'rd_restraints'
READ_RESTRAINTS FILE = CSRFILE
# This is an additional user provided file:
READ_RESTRAINTS FILE = 'nmr_2igh.rsr'
RETURN
END_SUBROUTINE
---------------------------------------------------
Prepare the "nmr_2igh.rsr" file in the following format:
MODELLER12 VERSION: USER FORMAT
R 3 1 1 28 2 2 0 2.9000 0.2000 HN:52 HB#:52
R 3 1 1 28 2 2 0 3.4000 0.2000 HN:49 HA:59
R 3 1 1 28 2 2 0 2.9000 0.2000 HN:48 HB#:48
R 3 1 1 28 2 2 0 2.9000 0.2000 HN:31 HB:30
R 3 1 1 28 2 2 0 4.0000 0.2000 HN:59 HG##:59
R 3 1 1 28 2 2 0 3.4000 0.2000 HN:25 HB#:24
R 3 1 1 28 2 2 0 3.4000 0.2000 HN:60 HG##:59
......
Hope it helps.
-Dong
http://compbio.ornl.gov/~nxy
Prevost Martine wrote:
> Hello,
>
> I would like to introduce a restrained distance between two atoms when
> generating a model with Modeller. Unfortunately when the two atoms for
> which I would like to retrain the distance are specified in my top file,
> the program seems to interpret this as all atoms between these two are
> concerned by possible restraints ?? I believe that when one wants to
> introduce restraints from NMR data that's what one wishes to do, i.e.
> specify pairs of atoms (one or many more) for which the distance should be
> within a certain range. Could somebody explain to me how the Modeller
> input should be modified so as to introduce such restraints ? Thank you
> very much.
>
> Martine.
>
>
> --------------------------------------------
> Martine Prevost
> Ingenierie Biomoleculaire, CP 165/64
> Universite Libre de Bruxelles
> 50 avenue Franklin Roosevelt
> B-1050 Bruxelles
> Belgium
> Phone : 32-2-650 2049
> Fax : 32-2-650 3606
> E-mail:
> --------------------------------------------