RE: Re: AMBER: ANAL/MMPBSA

From: Wu Yingliang <wuyliang.public.wh.hb.cn>
Date: Thu, 04 Dec 2003 15:53:25 +0800

Dear Prof. David A. Case,

Thank you very much for your reply so soon! After careful investigation for my work, the question still puzzles me now. My work is simply described as following: I ran a MD with GB (IGB=2) continnum medium for a docked protein complex by constraining all heavy atoms to heavy atoms in backbone (total running time: 50 ps). Then I carried out energy analysis by ANAL for final complex.restrt file, and by MMPBSA for obtained 100 snapshots collected by GB-MD from the equilibration of last. Corresponding input and output files are attached here:

1. file: anal.in

 decompose energies of protein complex
    1 0 0 0 0 1
    0 0.0 0.0 0.0 0.0
    1 0 0 0 0 0
  999.0 2.0 1.2 1.0
    1 10.0 10.0 10.0 10.0 10.0 10.0 10.0 10.0 10.0
ENERGY
Receptor
RES 1 388
END
Ligand
RES 389 419
END
END
STOP

2. part of anal.out

         VDW (N-B + 1-4) INTERACTION ENERGY MATRIX




            1 2
    1 -1642.731 -90.957
    2 -90.957 -88.750

         ELECTROSTATIC (N-B + 1-4) INTERACTION ENERGY MATRIX

             1 2
    1 -4085.362 -97.358
    2 -97.358 -292.527

 
          TOTAL INTERACTION ENERGY MATRIX

  
            1 2
    1 2044.961 -188.315
    2 -188.315 256.848


     TOTAL 2301.81 -188.31 2113.49


3. mm_pbsa.in


#
# Input parameters for mm_pbsa.pl
#
################################################################################
.GENERAL
#
# General parameters
# 0: means NO; >0: means YES
#
# mm_pbsa allows to calculate (absolute) free energies for one molecular
# species or a free energy difference according to Receptor + Ligand = Complex,
# DeltaG = G(Complex) - G(Receptor) - G(Ligand).
#
# PREFIX - To the prefix, "{_com, _rec, _lig}.crd.Number" is added during
# generation of snapshots as well as during mm_pbsa calculations.
# PATH - Specifies the location where to store or get snapshots.
# COMPLEX - Set to 1 if free energy difference is calculated.
# RECEPTOR - Set to 1 if either (absolute) free energy or free energy difference
# are calculated.
# LIGAND - Set to 1 if free energy difference is calculated.
#
# COMPT - Name of the parmtop file for the complex (not necessary for option GC).
# RECPT - Name of the parmtop file for the receptor (not necessary for option GC).
# LIGPT - Name of the parmtop file for the ligand (not necessary for option GC).
#
# GC - Snapshots are generated from trajectories (see below).
# AS - Residues are mutated during generation of snapshots from trajectories.
# DC - Decompose the free energies in single contributions (only works with MM and GB yet).
#
# MM - Calculation of gasphase energies using sander.
# GB - Calculation of desolvation free energies using the GB models in sander (see below).
# PB - Calculation of desolvation free energies using delphi (see below).
# MS - Calculation of nonpolar contributions to desolvation using molsurf (see below).
# If MS == 0, nonpolar contributions are calculated with the LCPO method in sander.
# NM - Calculation of entropies with nmode.
#
PREFIX Complex
PATH ../01_GenerateSnapshots/
#
COMPLEX 1
RECEPTOR 1
LIGAND 1
#
COMPT ../complex.top
RECPT ../receptor.top
LIGPT ../ligand.top
#
GC 0
AS 0
DC 0
#
MM 1
GB 1
PB 0
MS 0
#
NM 0
#
################################################################################
.GB
#
# GB parameters
#
# The first group of the following parameters are passed to sander.
# For further details see the sander documentation.
#
# IGB - Switches between Tsui's GB (1), Onufriev's GB (2),
# Jayaram's et al. GB (3) or Jayaram's et al. MGB (4).
# SALTCON - Concentration (in M) of 1-1 mobile counterions in solution.
# EXTDIEL - Dielectricity constant for the surrounding solvent.
#
# SURFTEN / SURFOFF - Values used to compute the nonpolar contribution Gnp to the
# desolvation according to Gnp = SURFTEN * SASA + SURFOFF.
#
IGB 2
SALTCON 0.00
EXTDIEL 80.0
#
SURFTEN 0.0072
SURFOFF 0.00
#
################################################################################

4. output of MM-PBSA


# COMPLEX RECEPTOR LIGAND
# ----------------------- ----------------------- -----------------------
# MEAN STD MEAN STD MEAN STD
# ======================= ======================= =======================
ELE -9444.69 66.76 -707.23 14.35 -8524.63 61.21
VDW -1785.20 28.03 -81.76 6.83 -1602.80 27.04
INT 8353.99 43.16 629.97 15.49 7724.02 43.57
GAS -2875.91 68.39 -159.01 14.68 -2403.41 65.05
GBSUR 166.13 2.29 19.71 0.32 159.14 2.23
GB -4762.94 57.56 -698.19 13.44 -4344.69 51.96
GBSOL -4596.80 56.50 -678.47 13.35 -4185.55 50.88
GBELE -14207.63 29.59 -1405.42 6.46 -12869.32 27.59
GBTOT -7472.71 41.33 -837.49 13.48 -6588.95 42.77

# DELTA
# -----------------------
# MEAN STD
# =======================
ELE -212.84 19.37
VDW -100.65 4.99
INT 0.00 0.00
GAS -313.49 20.89
GBSUR -12.72 0.51
GB 279.94 18.81
GBSOL 267.22 18.56
GBELE 67.10 5.06
GBTOT -46.27 4.67


As can be seen from above results, the electrostatic terms between ANAL and MMPBSA are very different. Such difference also happens in my other simulated protein complexes. Simultaneously, I also tested other combination of IDIEL and DIELC values, and always could not get similar results of electrostatic energy. Here, would you like to tell me how to reasonably design value of DIELC in ANAL input file?

Expect your help once more, and thanks a lot once more in advance!


Best wishes,

Wu Yingliang

=================================================================================
>On Wed, Dec 03, 2003, Wu Yingliang wrote:
>>
>> Then, how to understand the difference of
>> electrostatic energy between ANAL and MMPBSA?
>>


>You will have to be explicit about what differences you are referring to.
>Look at the test cases in amber7/test/vac_rna to see an example of comparing
>anal output to sander (or mm_pbsa) energies. The energies in analout.save
>and mdout.min.save are the same (look for "INITIAL ENERGY COMPONENTS" in
>the anal output). You can try to make similar comparisons for your system.


>..dac
                                                                                        


>
>==================================================================
>David A. Case | e-mail: case_at_scripps.edu
>Dept. of Molecular Biology, TPC15 | fax: +1-858-784-8896
>The Scripps Research Institute | phone: +1-858-784-9768
>10550 N. Torrey Pines Rd. | home page:
>La Jolla CA 92037 USA | http://www.scripps.edu/case
==================================================================
>-----------------------------------------------------------------------



-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Jan 14 2004 - 15:53:04 PST
Custom Search