Running a Go model with the AMHGMIN

GMIN can run various types of Go models with the AMH framework. This is useful for creating starting points for OPTIM, and also there are different forms of Go-like potentials that maybe interesting. To use this, first update and compile AMHGMIN as described in the GMIN Makefile.

Flags needed for GMIN input

The parameters behind the basin-hopping can be defined in the data file using the terms:

To begin with the AMH need to be turned on:

AMH

The interval of structures which are to be written to the file movie is termined by:

NINT_AMH    5

For structure prediction runs look something like this:

SLOPPYCONV  1.0D-3
TIGHTCONV   1.0D-3
UPDATES     5000
AMH
NINT_AMH    5
STEP  1.0 
MAXBFGS   1.0
TEMPERATURE 10.0

For slowly finding the lowest energy structure which is necessary as starting points for PATHSAMPLE runs shorter length steps, and much lower temperatures are necessary. The file then looks something like this:

SLOPPYCONV  1.0D-5
TIGHTCONV   1.0D-5
UPDATES     5000
AMH
NINT_AMH    5
STEP  0.4 
MAXBFGS   0.4
TEMPERATURE 1.0

Input file forGMIN

Two new terms have been included for use with the implicit membrane.

Firstly, the coordinates of the centre of mass of the system before the initial quench can be specified using:

SETCENTRE x y z

Specifying x y z as 0.0, 0.0, 0.0 will set the centre of mass at the origin and hence the centre of the membrane. A protein can be moved out of the membrane by altering the z coordinate.

Secondly, the system can be translated so that the centre-of-mass lies at the origin after every quench, using the keyword CENTRE. At present, CENTRE causes convergence issues as moving back to z=0 actually changes the energy. It is advised that CENTREXY is used instead, moving the protein back to (0,0,z) i.e. preserving the z-coordinate, by using the keyword:

CENTREXY

This solves the convergence problem mentioned above when using CENTRE.

CHARMM keywords

The toppar parameters that should be used are the EEF1.1.inp files. This is done using:

open read unit 11 card name toph19_eef1.1.inp
read rtf card unit 11
close unit 11

open read unit 12 card name param19_eef1.1.inp
read para card unit 12
close unit 12

The charmm keywords are as follows:

eef1 setup membrane slvt water slv2 chex nsmth 10 width 26.0 temp 298.15 -
unit 93 name "/home/jwll2/svn/CHARMM31/toppar/solvpar.1.inp" aemp 0.85
!gouy anfr 0.3 area 70. conc 0.1 offset 3.0 valence 1
update ctonnb 7. ctofnb 9. cutnb 15. group rdie

membrane introduces the membrane model and slvt water and slv2 chex specify that the exterior solvent is water and that the interior solvent is cyclohexane. nsmith, set to 10 as a default, determines how steep the transition is at the interface between interior and exterior. The width of the membrane can be altered by altering the value 26.0. This refers to the width of the membrane in angstroms and usually has a value 25-30. The last keyword aemp, default 0.85, determines the extent of strengthening of electrostatic interactions in the membrane (the smaller, the stronger). This parameter was empirically adjusted to give reasonable membrane insertion energies for model systems.

Uncommenting the commented line and adding a continuation to the line above allows the inclusion of Gouy Chapman theory adjustments to the membrane. This describes the effect of a static surface charge on the membrane potential. anfr describes the molar fraction of anionic lipids (e.g. a 70/30 mixture of PC/PG corresponds to ANFR 0.3, which is the default). area is the area (Angstrom^2) per lipid (default 70) and offset is the distance of the plane of negative charge, usually the phosphates, from the hydrocarbon/water boundary (default 3). conc and valence is the molarity and valence of the salt (default 0.1 and 1, respectively).