Hallo Amberries,
I would like to do the MM-PBSA analysis on single structures of a complex
(derived from a docking attempt). First of all - are there examples
available for using mm_pbsa on single structures (for instance, as
described in the recent Structure paper from Lee and Kollman)?
OK, here is a short description of what I did and where I failed (it's my
first attempt to use mm-pbsa, so I might have overlooked something very
basic):
I created topology files of the complex as well as the seperated receptor
and ligand structure with tleap and provided them as parm_complex.top,
parm_receptor.top and so on. I minimized the complex and provided the
result as soln1_min.rst.gz (belonging to parm_complex.top). I then adapted
the anal.in file from mm_pbsa/EXAMPLE1 (just changing the residue number)
and created anal_complex.in, anal_receptor.in etc. Next I copied over the
missing parameter files (*crg, *siz, atmtypenumbers) from EXAMPLE1. Then I
edited the mm_pbsa script. I had to change to bash as ksh is not running
on my Linux system and had to comment out the code about setting the zcat
variable (but added zcat='/bin/zcat' manually) plus some minor changes.
Now I set all the program locations, adjusted TRAJECTORY, NTOTAL, LSTART
etc.and -- the program breaks telling me:
fmt: read unexpected character
apparent state: unit 5 (unnamed)
last format: (A4,2X,I5,1X,2A4,2X,I4,4X,3F8.3,3F8.4,I5)
lately reading sequential formatted external IO
./mm_pbsa: line 926: 25185 Aborted
"$GB_program" <"$pdbname""$number".GB >"$pdbname""$number".GB.out
The problem seems to start with the file test1__complex.crd.1 which is
created by the script. It doesn't fit the topology file at all. mm_pbsa
then calls
anal -O -i anal_complex.in -o anal_complex.1.out -c test1__complex.crd.1
-p parm_complex.top -p1 test1__complex.pdb.1.tmp
anal_complex.1.out then of course contains energy values like "nan" and
"-inf" which is probably causing the described halt of the script. By
the way, anal reports a perfect result when giving it the original
soln1_min.rst file.
Any hints are most welcome!
Thanks in advance
Raik
=====================================================
Raik Grünberg | Bioinformatique Structurale
| Institut Pasteur
| 25-28 rue du docteur Roux
raik_at_pasteur.fr | 75015 Paris
Tel: +33/1.45.68.87.37 | France
-----------------------------------------------------
http://www.raiks.de/contact.html
=====================================================
Received on Wed Nov 14 2001 - 06:45:38 PST
