Loading MTZ data - volume is restricted to a small sphere within the protein model
Hi folks, I recently installed Clipper, and on loading some example PDB and MTZ data, I'm seeing just a sphere of density data that does not cover the full protein model. This behaviour is not restricted to a single example. Is it expected?
The example files I used are in 6DIL https://www.rcsb.org/structure/6DIL. I downloaded "PDB Format" and "Map Coefficients (MTZ Format)" from the drop-down menu. Loading the PDB by drag-and-drop into ChimeraX works as expected (coordinates do show). On loading the MTZ by drag-and-drop, and selecting the PDB model to import data into, I see the below representation. The density does not cover the full set of protein coordinates - it's limited to a sphere within the protein.
Curiously, I saw this on coot too - so maybe I'm just getting the options incorrect? I did try shifting the cutoffs in the "Volume Viewer" Tool, and the mesh does change but it's still limited to this sphere. Changing from a mesh surface to "Maximum" shows that the available data is limited to a cube that does not cover the protein. Is there a way to view the full density?
Out of interest, I also tried generating the mesh myself, but only got halfway. This at least shows that the data does cover the whole protein:
import gemmi import numpy as np import meshplot as mp from skimage.measure import marching_cubes #run 'gemmi mtz2cif 6dil_phases.mtz 6dil_phases.cif' first doc = gemmi.cif.read('/Users/ljmartin/Desktop/6dil_phases.cif') rblocks = gemmi.as_refln_blocks(doc) rblock = rblocks[0] size = rblock.get_size_for_hkl(sample_rate=2.6) full = rblock.get_f_phi_on_grid('pdbx_FWT', 'pdbx_PHWT', size) array = np.array(full, copy=False) complex_map = np.fft.ifftn(array.conj()) scale_factor = complex_map.size / full.unit_cell.volume real_map = np.real(complex_map) * scale_factor v,f,n,_ = marching_cubes(real_map, 0.05) mp.plot(v, f)
Hi Lewis,
This default map mode in the Clipper plugin is (like Coot) designed that way for two main reasons:
(a) for a crystallographic dataset there is no one definition of what constitutes the “full map” - it just goes on forever in all directions.
(b) the main use case is for investigating the details of the model in the context of the map - when zoomed in to the scale of individual residues, keeping the whole map drawn just adds lots of unnecessary overhead.
Anyway, here’s the basic overview of what you can do:
- pan around with middle-click-and-drag - the sphere of density will follow where you go;
- increase/decrease the size of the sphere witb the command “clipper spotlight radius {value}”;
- expand the map and mask it to cover any arbitrary selection within your model with “clipper isolate {sel}” (return to the sphere mode with “clipper spotlight”);
- save the currently displayed map region to mrc format using the save command in exactly the same way you’d save a standard ChimeraX Volume.
Hope this helps!
— Tristan
On Wed, 15 Feb 2023 at 01:26, Lewis James martin via ChimeraX-users < chimerax-users@cgl.ucsf.edu> wrote:
Hi folks, I recently installed Clipper, and on loading some example PDB and MTZ data, I'm seeing just a sphere of density data that does not cover the full protein model. This behaviour is not restricted to a single example. Is it expected?
The example files I used are in 6DIL https://www.rcsb.org/structure/6DIL. I downloaded "PDB Format" and "Map Coefficients (MTZ Format)" from the drop-down menu. Loading the PDB by drag-and-drop into ChimeraX works as expected (coordinates do show). On loading the MTZ by drag-and-drop, and selecting the PDB model to import data into, I see the below representation. The density does not cover the full set of protein coordinates - it's limited to a sphere within the protein.
Curiously, I saw this on coot too - so maybe I'm just getting the options incorrect? I did try shifting the cutoffs in the "Volume Viewer" Tool, and the mesh does change but it's still limited to this sphere. Changing from a mesh surface to "Maximum" shows that the available data is limited to a *cube *that does not cover the protein. Is there a way to view the full density?
Out of interest, I also tried generating the mesh myself, but only got halfway. This at least shows that the data does cover the whole protein:
import gemmi import numpy as np import meshplot as mp from skimage.measure import marching_cubes
#run 'gemmi mtz2cif 6dil_phases.mtz 6dil_phases.cif' first
doc = gemmi.cif.read('/Users/ljmartin/Desktop/6dil_phases.cif') rblocks = gemmi.as_refln_blocks(doc) rblock = rblocks[0]
size = rblock.get_size_for_hkl(sample_rate=2.6) full = rblock.get_f_phi_on_grid('pdbx_FWT', 'pdbx_PHWT', size) array = np.array(full, copy=False) complex_map = np.fft.ifftn(array.conj()) scale_factor = complex_map.size / full.unit_cell.volume real_map = np.real(complex_map) * scale_factor
v,f,n,_ = marching_cubes(real_map, 0.05)
mp.plot(v, f)
[image: Screenshot 2023-02-15 at 11.53.56 am.png] _______________________________________________ ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
Wonderful, thanks Tristan - it works well! Resolved. Lewis
On February 16, 2023, Tristan Croll tcroll@altoslabs.com wrote:
Hi Lewis, This default map mode in the Clipper plugin is (like Coot) designed that way for two main reasons: (a) for a crystallographic dataset there is no one definition of what constitutes the “full map” - it just goes on forever in all directions. (b) the main use case is for investigating the details of the model in the context of the map - when zoomed in to the scale of individual residues, keeping the whole map drawn just adds lots of unnecessary overhead. Anyway, here’s the basic overview of what you can do:
- pan around with middle-click-and-drag - the sphere of density will
follow where you go;
- increase/decrease the size of the sphere witb the command “clipper
spotlight radius {value}”;
- expand the map and mask it to cover any arbitrary selection within
your model with “clipper isolate {sel}” (return to the sphere mode with “clipper spotlight”);
- save the currently displayed map region to mrc format using the save
command in exactly the same way you’d save a standard ChimeraX Volume. Hope this helps! — Tristan On Wed, 15 Feb 2023 at 01:26, Lewis James martin via ChimeraX-users <chimerax-users@cgl.ucsf.edu mailto:chimerax-users@cgl.ucsf.edu> wrote:
Hi folks, I recently installed Clipper, and on loading some example PDB and MTZ data, I'm seeing just a sphere of density data that does not cover the full protein model. This behaviour is not restricted to a single example. Is it expected?
The example files I used are in 6DIL https://www.rcsb.org/structure/6DIL. I downloaded "PDB Format" and "Map Coefficients (MTZ Format)" from the drop-down menu. Loading the PDB by drag-and-drop into ChimeraX works as expected (coordinates do show). On loading the MTZ by drag-and-drop, and selecting the PDB model to import data into, I see the below representation. The density does not cover the full set of protein coordinates - it's limited to a sphere within the protein.
Curiously, I saw this on coot too - so maybe I'm just getting the options incorrect? I did try shifting the cutoffs in the "Volume Viewer" Tool, and the mesh does change but it's still limited to this sphere. Changing from a mesh surface to "Maximum" shows that the available data is limited to a cube that does not cover the protein. Is there a way to view the full density?
Out of interest, I also tried generating the mesh myself, but only got halfway. This at least shows that the data does cover the whole protein:
ChimeraX-users mailing list ChimeraX-users@cgl.ucsf.edu mailto:ChimeraX-users@cgl.ucsf.edu Manage subscription: https://www.rbvi.ucsf.edu/mailman/listinfo/chimerax-users
participants (2)
-
Lewis James martin -
Tristan Croll