makeGrid

makeGrid takes four inputs: a control file, an AMBER topology file, an AMBER trajectory, and a reference structure in AMBER restart format. The program then calculates the test particle grids (van der Waals and electrostatic) for a particle and location specified in the control file. These grids are calculated based on the input topology and trajectory, and projected onto the specified coordinates relative to the provided restart file. makeGrid is designed as a standard Unix program; the input trajectory should be piped to the makeGrid command and other inputs specified on the command line:

cat | makeGrid

Control file for makeGrid. The control file consists of a standard AMBER-style namelist followed by atom specifications, described in detail below. The namelist specifies control variables for the run (cutoff, grid dimensions and spacing) as well as indicating the atoms that define the grid center and axes. The subsequent atom specifications indicate which parts of the system to include in the free energy calculation. It is common in a protein-ligand complex to include everything _except_ the ligand in the calculation. This way the grid reflects only the influence of the protein and is not biased by nearby atoms of the ligand.

AMBER format topology file for makeGrid. This should be the correct topology for both the input trajectory and coordinate files.

Reference coordinates in AMBER restart format. Once the grid is calculated, it will be rotated and translated to fit the appropriate atoms of this particular coordinate set.

Specifies location of lennard-jones particle(s) to be added to all calculations. The intent of this feature is to allow the user to calculate test partical grids for the case where a lennard-jones particle is also added at the specified locations. Can be usefull for modeling a simple modification and calculating a new test partical grid without running an additional simulation. This should be used with caution, as the results are subject to more noise then a standard calculation.

van der Waals "cost field"; the vdw file is a scalar field of the free energy required to insert the specified test particle on a grid of points relative to the ligand or atoms of interest, based on the coordinates and parameters supplied in the and files. The grid is specified relative to the coordinates of the specified atoms as found in the file. See the Field section below for format information.

weighted electrostatic potential field; the esp file is a scalar field in a format similar to the vdw file. However, it contains the average electrostatic potential at each grid point. This is not a true electrostatic potential -- instead, it is weighted by the probability that the test particle would occupy this location (boltzman factor of the the vdw grid). This weighting avoids the infinite electrostatic contributions that result from overlaps of the test charge with real atoms. Again, format information is supplied with the description of the Field delegate, below.

indicates the location of lennard-jones particles added using the ljp file (above). This is intended for use with MIDAS but is not critical.

detailed save file; this file contains all of the makeGrid output data in a format suitable for more detailed analysis. In particular, the sav file contains the data necessary to concatenate the output of makeGrid runs on two different trajectories of the same topology.

makeGrid input format

Here is a sample input file:

&cntrl
Rprobe = 1.9080, Eprobe = 0.1094,
iAtomO = 3877, iAtomX = 3879, iAtomY = 3883,
nGridX = 15, nGridY = 15, nGridZ = 15,
sGridX = 0.50, sGridY = 0.50, sGridZ = 0.50,
Xtrans = 0.0, Ytrans = 0.0, Ztrans = 0.0,
Temp = 300.0, cutoff = 8.0,
&end
include all
exclude residueName DMT

The namelist is delimited by the " &cntrl" and " &end" lines. It defines the input variables for the run:

Rprobe, Eprobe

van der Waals parameters for the test particle. Rprobe is the R* value (in Angstroms) and Eprobe the epsilon (kcal/mol).

iAtomO

Origin atom for the grid. In the absence of other input, the grid will be centered on this atom.

iAtomX

Atom to define the x-coordinate of the grid.

iAtomY

Atom to define the y-coordinate of the grid. iAtomO, iAtomX, and iAtomY are used to define a constant frame of reference for the grid. The z- axis is generated by the right hand rule. Typically all three are ligand atoms near the region of interest. Also, these atoms should not move much relative to one another, since they are assumed to be a relatively constant frame of reference.

nGridX, nGridY, nGridZ

number of grid points in each dimension; since the grid point (0,0,0) is centered on the origin (iAtomO), this must be an odd number.

sGridX, sGridY, sGridZ

grid spacing in each dimension, in Angstrom. A grid spacing of 0.5 or 0.25 Angstrom is typical.

Xtrans, Ytrans, Ztrans

translation of the grid away from iAtomO, in Angstroms. This can be used to center the grid at a position other than iAtomO (anywhere in space).

Temp

Temperature used for the Boltzmann weighting and free energy calculation, in Kelvin. 300K is typical.

cutoff

the cutoff (in Angstroms) used to truncate calculation of the van der Waals and electrostatic interactions. Note that this does _not_ have to be the cutoff used to generate the input trajectory.

The include and exclude commands specify the atoms to consider in the test particle insertion and electrostatics calculations. Typically all atoms are included, then the ligand of interest is specifically excluded. The commands are:

include/exclude all
residueName
residueNumber
residueRange
atomNumber
atomRange

where , , , etc. refer to residue names and residue or atom numbers as specified in the input topology