Hi,
Also, trying to use the command
> gistpp –i gist-gO.dx –i2 complex.pdb –op defbp –opt cutoff1 10 –o
> perturbedcomplex.dx
>
> the error "The ligand structure file is expected in i2 NOT in i!!
> terminate called after throwing an instance of 'std::bad_alloc'
> what(): std::bad_alloc
> Aborted"
>
This is indeed a strange result. It may help to use the latest version of
gistpp, which can be downloaded here:
https://github.com/gosldorf/gist-post-processing. One benefit of the
up-to-date version is that it has better error handling in place and may
provide a better clue to what went wrong when you executed that command.
The message you received starts with a warning "The ligand structure file
is expected in i2 NOT in i!!" which was set to output anytime the operation
defbp was called and so should be disregarded if option use is correct --
as it appears to be in your example. The what: std::bad_alloc error can
mean multiple things, so the best I can do is suggest a few issues that
could be the root of the problem:
1) If either of the provided input files are not found the older version of
gistpp will throw a similar error, rather than a file not found error (this
is fixed in the latest version).
2) This error message may occur when gistpp has run out of system memory to
access while running (though this is unlikely the case).
3) Finally, the cutoff defined in the command is a distance in angstroms
and is not usually set as high as 10, since the GIST regions are defined in
number of voxels (at side length 0.5 angstrom) the value of 10 here would
correspond to all voxels that are within a distance of 20 voxels from each
heavy atom in complex.pdb. That may be too large a volume for the code to
handle or may be a volume that exceeds the size of the original gist region
of interest, meaning the size of the dx maps in use.
If it turns out that issues 1+2 are not the problem it would be very
helpful if you could send me the gist-gO.dx and complex.pdb files in use
here off of the mailing list (directly to: vpsramsey.gmail.com) so that I
can get to the bottom of the issue and fix the bug, if one is present.
Could you please provide some more details on how to produce the contours
> figure presented in AMBER Advanced Tutorial 25 - Introduction page?
>
I split it up my answer to this into two parts: first general tips for
rendering gist dx maps in software like VMD or PyMol, then how to
specifically obtain the data being rendered in the example image for your
system of interest.
To visualize gist energy surfaces around a macromolecule (such as CB7 in
the example, or a protein binding site) you can load the gist .dx maps into
VMD or PyMol (and other similar visualization software) along with a
structure of the macromolecule. The .dx maps are by default considered as
an isosurface in these software and can be contoured at any isovalue (the
value that defines the interface that the surface represents). To produce
high quality .dx map surfaces I would recommend setting the software to
render the isosurface as a surface (rather than wiremesh or points) and
transparent so that the macromolecule is also visible.
The contour maps around CB7 in the GIST tutorial introduction page were
made using a .dx map of normalized solute-water energies, where normalized
means that the energies are reported in per water units (kcal/mol/water).
This .dx map is not output by default by GIST, but is calculated and stored
in the spreadsheet output text file (default name: gist-output.dat or
specified with the keyword 'out' in the GIST command). To extract this data
into a .dx map we can use a gistpp command:
gistpp -i gist-output.dat -i2 gist-gO.dx -op makedx -opt const 14 -o
gist-Esw-norm.dx
Where gist-output.dat is the spreadsheet file mentioned above, gist-gO.dx
is the oxygen density dx map (provided here solely as a template), makedx
is the operation being used, the file being output is gist-Esw-norm.dx, and
-opt const 14 instructs gistpp to make the 14th column (count starting at
0) from the gist spreadsheet output into a dx map. It might be the case
that in older versions of GIST (I used AmberTools17 here) the column
numbers may have changed, so it is best to confirm from the header of the
spreadsheet file before running this command.
I believe that the isovalues shown in the figure, -8.5 kcal/mol/water and
-4.0 kcal/mol/water, are energy quantities relative to bulk water (-9.53 in
tip3p) so if we were to use the newly created gist-Esw-norm.dx file to
render the same surfaces we would make two surface representations, one
where the isovalue is set to -18.03 (orange in original figure,
kcal/mol/water) and -13.53 (blue in original figure, kcal/mol/water).
Sorry that this response ended up being long, hope this helps,
--Steve
On Tue, Feb 6, 2018 at 7:14 PM, Sofia Vasilakaki <svasilak.chem.uoa.gr>
wrote:
> Hello,
>
> Could you please provide some more details on how to produce the contours
> figure presented in AMBER Advanced Tutorial 25 - Introduction page?
>
> Also, trying to use the command
> gistpp –i gist-gO.dx –i2 complex.pdb –op defbp –opt cutoff1 10 –o
> perturbedcomplex.dx
>
> the error "The ligand structure file is expected in i2 NOT in i!!
> terminate called after throwing an instance of 'std::bad_alloc'
> what(): std::bad_alloc
> Aborted"
>
> was produced. What this means?
>
> Thanks,
> Sofia
>
>
>
>
>
>
>
>
> _______________________________________________
> 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 Tue Feb 06 2018 - 18:00:03 PST
