AMBER Archive (2002)

Subject: help/suggestions needed with PROFEC

From: Salinthip Thipayang (salinthip.thipayang_at_ic.ac.uk)
Date: Fri May 03 2002 - 10:03:47 CDT

Dear AMBER users,

If anyone has any idea or has any experiences in using PROFEC please take a
minute to look at this email. I posted an email to AMBER mail reflector
archivebefore ( which is the email that I also attach at the end of this
message) but I have got no reply at all. So if there are still PROFEC users out
there, even if
you cannot solve my problems, could you please just reply so that I know
there are still some people using PROFEC like me.

I have 2 questions that I really need the suggestions/answers/recommendations
from you.

My first question is when I run this command line:

cat <.crd file> | /usr/local/amber6/exe/profec_grid <makegrid.in> <.prm file>
<.rst file>

which (in my understanding) is equivalent to the command line given inPROFEC
manual:

cat <traj> | makeGrid <in> <top> <crd> <ljp> <vdw> <esp> <obj> <sav>

where <.crd file> is my coordinate file from MD run using SANDER6

               <makegrid.in> is the input file as given in the example in the
manual
               <.prm file> is the parm topology file created by XLEAP
               <.rst file> is the restart file after the MD run (same MD run
that produced the <.crd file> specified before) using SANDER6

Note that there is no makeGrid program, instead it is profec_grid in the
/usr/local/amber6/exe/ directory. This is exactly the same quesion that Anna
Ferrari
posted in the mailing list on 02/03/2001.
  

I got the following error message:

        3 1
makeGrid in top crd ljp vdw esp obj sav

( I am not sure if the numbers 3 and 1 above the actual message is refering to
the positions of the makegrid.in file and <.rst file> from my command line)

>From my understanding I think I only need to put 4 inputs i.e. 1) the <.crd file> 2) makegrid.in 3) <.prm file> and 4) <.rst file> in the profec_grid command line
in order to get the outputs <lpj>, <vdw>, <esp>, <obj> and <sav> out. These
output files will then be used in makeDiffGrid and Field (with Midas) to
qulitatively
calculate the ddGbind and to view the grid respectively. Please correct me if I
understand wrongly about this.

My second question (which is not really valid at this stage unless I can get the
makeGrid command to work first) is that how would one define the R* (van der
Waals radius) and the epsilon (van der Waals well depth) used in the makegrid.in
as Rprobe and Eprobe values respectively? Say if I want to change the COO-
(carboxyl) group in GLU residue in the native enzyme to other group such as
CH3, NH2 etc. in the mutant enzyme, what are the values of R* and epsilon to use
here? Do I use the atomic carbon van der Waal parameters for atom type C of the
carboxyl or carbonyl carbon (1.9080, 0.0860 as in parm98.dat file) or do i
need to use values for the united atom model for COO- group? The example of the
makegrid.in file in the PROFEC manual seems to use Rprobe= 1.9080 and
Eprobe= 0.1094 (taken from parm98.dat) which are for the sp3 carbon atom typeCT;
whereas the two references that I quoted in my previous email (L. Wang
et al 1998 and R.J. Radmer and P.A. Kollman 1998) seem to use the united CH3
atom with values of R*= 2.0 A and epsilon=0.15 kcal/mol. These values seem to be
taken from table XIX on page 781 of S.J. einer et al, J. Am Chem. Soc.,
1984, 106. 765-784 which shows the R* and epsilon values for nonbonded
parameters.
  

Thank you very much for spending some of your time reading/ replying this email.
Your help/ suggestions/ recommendations are really appreciated. Thank you
very much again.

Yours Sincerely,

Salinthip Thipayang
  
  
  

The following is my previous email that I posted on 19/04/2002 Dear PROFEC
users, I am trying to use PROFEC to qualitatively predict the effect of changing
the residue in the binding site of a protein from the free energy calculations
btained from PROFEC. The first step of using PROFEC ( as I understand from the
manual) is to make particle grids van der Waals and electrostatic) for the
protein-ligand and for ligand-water systems, using the makeGrid file (or
profec_grid
in AMBER6). The manual describes the command line as:

cat <traj> | makeGrid <in> <top> <crd> <ljp> <vdw> <esp> <obj> <sav>

I understand that cat <traj> is obtained from the mdcrd file as obtained by
SANDER. makeGrid <in> is the input command file which is briefly described and
an
example is given in the manual. <top> is the AMBER topology file as btained in
Xleap. <crd> is the mdcrd file contained coordinate references and can be
obtained by Xleap. But how and here can I obtain the <vdw>, <esp>, <obj> and
<sav> files? What programs do I have to use to get them? The manual
descibes what they are but does not say how are they calculated and what
programs required to get these files to generate the grids. I notice the second
step
after the makeGrid process is the makeDiffGrid and we need to generate the
<protein.vdw>, <solvent.vdw>, <output.vdw>, <protein.esp> etc. Anther obvious
question is how can one generate these <protein.vdw>, <solvent.vdw>?

I have read the quoted references about PROFEC e.g. from

R.J.Radmer and P.A.Kollman, 'The application of three approximate free energy
calculations methods to structure based ligand design: Trypsin and its complex
with inhibitors', J. Computer-Aided Mol. Design, 12: 215-227, 1998.

L.Wang et.al., 'Can one predict protein stability? An attempt to do so for
Residue 133 of T4 Lysozyme using a combination of free energy derivatives,
PROFEC,
and free energy perturbation methods', Proteins: Structure, Function and
Genetics 32: 438-458, 1998.

M.A.L.Eriksson et.al., 'Prediction of the binding free energies of new TIBO-like
HIV-1 reverse transcriptase inhibitors using a combination of PROFEC,
PB/SA, CMC/MD, and free energy calculation', J. Med. Chem., 42: 868-881, 1999

These references are very useful but they do not describe the methods of using
PROFEC in terms of the computational set up details. If there are any other
references that you think may be useful, please quote them as well.

Any comments, any recommendation will be greatly appreciated. Thank you everyone
in advance for spending time reading/
replying this email.

I really hope to get some reply please!

Yours sincerely,

Salinthip Thipayang 

-- 
-------------------------------------------------------------------------------
Salinthip Thipayang (Miss)

PhD Research Student
Biological and Biophysical Chemistry Research Group
Chemistry Department
Imperial College of Science, Technology and Medicine
Exhibition Road
London SW7 2AY 
The United Kingdom
Tel: +44(0)207 5945851
Fax: +44(0)207 5945845
Email: salinthip.thipayang_at_ic.ac.uk
--------------------------------------------------------------------------------