Re: [AMBER] exact values corresponding to the grid points in GIST calculations

From: Steven Ramsey <vpsramsey.gmail.com>
Date: Sat, 9 Jan 2016 18:29:14 -0500

Hello Mohan!

There's a tool for post processing GIST, called GISTPP, which you can
download from the 4th section of the AMBER GIST tutorial
http://ambermd.org/tutorials/advanced/tutorial25/section4.htm.

GISTPP allows you to define voxels of interest and evaluate their values
(values here being any quantity assess during a GIST calculation), but
sometimes requires multiple operations to do so. It isn't entirely clear to
me exactly what's going on in the image provided, but I'll describe a
workflow I would use based on that image being a contour of Esw-dens > 0.5.

What I would do to determine the highlighted volumes coordinates would be
to create a binary map using clear1 (binary here refers to: voxels that
meet a criterion will contain a value of 1 and voxels that do not meet that
criterion will contain a value of 0):

gistpp -i gist-Esw-dens.dx -op clear1 -opt cutoff1 0.5 -opt gt1 -o
greaterthan0.5Esw.dx

This will create a binary map of all voxels that contain greater than 0.5
in the provided Esw-dens map and should match the image created by
contouring the original file, options cutoff1 define the threshold value
and gt1 that you're interested in greater than said threshold. From here
you can use the group command to separate each of the chunks you see in the
image based on spatial continuity:

gistpp -i greaterthan0.5Esw.dx -op group

This operation searches the resulting clear1 map and attempts to assign
each voxel with data in it to a group, keeping all neighboring voxels
together. What will happen is the highlighted section in your image will be
its own group and by viewing the produced mastergroup.pdb file you will be
able to visualize it, find its group number (stored as resid), and also
find the x y z coordinates of each voxel. In addition you can use these
group maps to evaluate other thermodynamic quantities of interest with the
GISTPP operation mult if desired.


Another option to solve for the coordinates is to convert the original
clear1 'greaterthan0.5Esw.dx' file I described using the operation printpdb:

gistpp -i greaterthan0.5Esw.dx -op printpdb -o greaterthan0.5Esw.pdb

This will create a pdb file containing every voxel in the above binary map
(wherein only voxels that fit a criterion are represented), however will
not separate out voxels based on locations or label them, therefore
requiring more searching on your part (potentially) in vmd.


There may be even more routes to determining the coordinates using the
operations available in GISTPP, but those are the two that I suspect will
work the smoothest.


Hope this helps, happy to be of further assistance,

--Steve

On Sat, Jan 9, 2016 at 5:36 AM, Mohan Pradhan <pradhanmohan56.gmail.com>
wrote:

> Dear Amber users,
>
> I have performed the GIST calculations around the ligand binding site of a
> protein. I am able to visualize the energetic/entropic values contoured on
> the 3D grid in VMD/Chimera.
>
> I now want to extract the exact values corresponding to the specific grid
> points/boxes.Basically, how do I know which X, Y, Z coordinates (columns 2,
> 3, 4 in the grid-output.dat file) correspond to which grid points when I
> visualize the structure with the contour map.
>
> Please find attached an image file for the same question.
>
> Many Thanks!
> Mohan
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Sat Jan 09 2016 - 15:30:03 PST
Custom Search