Usage questions: Please ask here first [0]

[hainm]

For any questions relating to how/why/ ... Or anything you don't know where to ask.

For bug, suggestion, ... please open seperate issue

[gph82]

Hi, I notice a bunch of tempfiles in my /tmp directory after using nglviewer, containing essentially the molecular coordinates in pdb format. Do they stay there because of abrupt termination of nglivew or something?

[hainm]

@gph82 I guess we did not do great job in cleaning those files. No, nglview do not need them to be there all the times.

updated: remove expired link.

[kain88-de]

Hi is there a way to access and manipulate the rotation and zoom independently in the view object. NGL seems to support this. nglviewer/ngl#240 (comment)

I know about view.orientation but it handles zoom and rotation and at the same time and not separately.

[hainm]

Hi, can you open a new issue so we can make it in our todo list. thx.

[lorenzoFabbri]

Hi all. I would like to do the following thing: each cell of my iPython notebook should contain only a "portion" of the entire protein. I explain: in each cell I would like to make visible only the portion of the protein that corresponds to a particular sequence. Is this possible? Is it possible to at least highlight in some way some portions of a protein using a sequence of amino acids as input? Thank you.

[hainm]

@lorenzoFabbri

: in each cell I would like to make visible only the portion of the protein that corresponds to a particular sequence. Is this possible?

No it's not possible. You can can always create 3 views and display them.

views = [nglview.show_structure_file('my.pdb') for _ in range(3)]

# views[0]: do something
views[0]

# another cell
# views[1]: do something
views[1]
...

Is it possible to at least highlight in some way some portions of a protein using a sequence of amino acids as input? Thank you.

May be just adding representations for given residue selection if you know where they are?
http://arose.github.io/ngl/api/tutorial-selection-language.html

PR is welcome if you think those utils useful for you and others. Cheers.
Ping @arose here in case you have different answers.

[kain88-de]

In a notebook I recently viewed two structures in different notebook cells. That worked fine.

[hainm]

In a notebook I recently viewed two structures in different notebook cells. That worked fine.

Yes, you can create as many views as you want.

[kbsezginel]

Hi, is there a way to assign color to a given atom type in ball and stick model? It seems like other than common atoms in organic chemistry the rest of the atoms are represented with white color which makes it hard to distinguish with hydrogen atoms. Is there a way use something like CPK coloring? Thanks!

[kain88-de]

Is there a way to install a development version as a isolated environment? I found using a different conda environment screws with the javascript in my root environment.

[hainm]

@kbsezginel can you please provide screenshot?

Ping @arose

[arose]

is there a way to assign color to a given atom type in ball and stick model

There is CPK coloring based on the element. So if everything is white there is likely an error in determining the element. I would need an example file to look into that.

[kbsezginel]

When I create a list of different atoms such as this:
atom_names = ['C', 'H', 'O', 'N', 'F', 'Cl', 'Br', 'Li', 'Zn', 'Cu', 'Pd', 'Co']
separate them 4 angstroms apart and save to a pdb file, this is what I see with nglview:

When I load the same pdb file using a molecular graphics software this is what I get:

As you can see only a handful of atoms are colored and their radii are all the same as well.

I attached the jupyter notebook and pdb file below:
atom-color.zip

[arose]

The issue are non all-upper-case atom names. Fixed in the ngl development version. Will take some time to go into nglview. Workaround is providing all-upper-case atom names.

[hainm]

Will take some time to go into nglview

if you push to npm, will be available tonight :D

[arose]

if you push to npm, will be available tonight :D

there are some issues with the current version... ;)

[kbsezginel]

Thanks a lot! It seems like the radii are all the same though, is that also fixed in the ngl development version?

[arose]

It seems like the radii are all the same though

that is just due to the default parameters of the ball+stick representation. Have a look at the radiusType parameter (http://arose.github.io/ngldev/api/global.html#StructureRepresentationParameters)

[kbsezginel]

Hi again! I run into this problem that in ball and stick model the bonds are not displayed after a certain number of atoms. Is there a way to force calculation of the bonds or does it just skip that when there is too many atoms?

Here with approximately 300 atoms:

Here with approximately 500 atoms:

Thanks!

[arose]

@kbsezginel Can you please provide the file you are loading. Thanks!

[kbsezginel]

Here, I attached the pdb files. Thanks!

n_atoms.zip

[arose]

You assign the same group/residue number to all molecules. If you give every molecule its own number the bond calculation will work fine.

[kbsezginel]

Thanks, that seems to get the bond calculation working however now I am seeing incorrect assignments of the bonds for some molecules:

When I visualize the same file without groups I see no bonds as expected:

I also tried with a smaller number of molecules:
Group:

No group:

Any idea why that's the case?
Thanks!

Attached pdb files here:
mol_grouping.zip

[arose]

I haven't looked at it but my hunch is that it will work if you give each group its own distinct three letter name instead of calling them all MOL. For efficiency reasons, NGL tries to reuse calculated bonds for identical groups. However the identity calculation seem wrong here.

[kbsezginel]

Thanks, that worked perfectly!

[morpholin]

I wanted to ask, if it is possible to also display external densities, extracted e.g. from a Gaussian Cube file?

[hainm]

This page is too long now. I close it and open another issue: #785

[sperezconesa]

Hello,

Is it possible to make selections based on bfactor or something similar like:

view.add_surface('protein and bfactor < 1.0')

Thank you very much and keep up the good work!
Best,
Sergio

[hainm]

hi @sperezconesa: I don't think so. Please see the selection language here: http://nglviewer.org/ngl/api/manual/selection-language.html

[DustBytes]

Need few "How to" ideas. I don't have any jupyter for demo :(

1. I have a table with two residues in each row. How to highlight/make bold etc these two residues in the nglviewport when hovering over the table rows. This working I saw in jsmol.
2. Other similar idea, (any working will do), how to highlight two or more residues when hovering on certain residues in nglviewport?
3. How to change the name of the residue/atom when we hover the cursor over the ngl viewport? Right now it looks like ATOM: [GLN]261:A.CA(1mnp.pdb) . How can we have cutom name?
4. How to colour each chain with "chainid" but also put cutom colours on certain resiudes.
   Bug opened here : Help: Colouring ngl#822 and here: Help: Colouring #937

[hainm]

hi @DustBytes: sorry to disappoint you again. Those type of questions should be addressed in NGL's repo: https://github.com/nglviewer/ngl/issues

[juliasubbotina]

• Question about usage? Please search your issue here first: Usage questions: Please ask here first [0] #589 and Usage questions about nglview: Please ask here first [1]  #785

• Version report

python -c 'import nglview; print(nglview.__version__)'
3.0.3
python -c 'import ipywidgets; print(ipywidgets.__version__)'
7.6.5

Hello,--
I started to experience the issue with shape function. The PDB has only 1 atom and the radii_core > 10.0. With the code below it does not produce the view. It only does so if the radii of the sphere is less than 10.

view=nv.show_mdanalysis(system)
view.add_representation(repr_type='spacefill')
view.shape.add_sphere([ 0.0, 0.0, 0.0 ], color_core, radii_core*10, "AUM")

However, the same view.shape.add_sphere() works just fine for systems with large number of atoms. it doesn't break at radii > 10.

view=nv.show_mdanalysis(system)
view.add_representation(repr_type='spacefill', selection='PEG',  opacity=0.5, color=color_shell, radius=0.5*10*float(r_poly))
view.update_representation(component=view.n_components - 1, opacity=0.9) 
view.update_representation()
view.shape.add_sphere([ 0.0, 0.0, 0.0 ], color_core, radii_core*10, "AUM")

Any idea why this can happen?

Thanks.

[LijieZhong]

• Version report

python -c 'import nglview; print(nglview.__version__)'
3.1.2
!python -c 'import ipywidgets; print(ipywidgets.__version__)'
8.1.2

from ase.io import read,write
import nglview as ngl
data = read("./)pro.gro")
vis = ngl.show_ase(data)
vis.use_style = 'ngl'
vis
![stru](https://github.com/nglviewer/nglview/assets/64005198/72b9546f-8d90-40e7-801d-4421b1ffd384)

Hello, I am trying to view a mixed solvent system using the above code, but the lithium ions in this system do not seem to display properly, how should I solve this problem
Thanks

[juliasubbotina]

Not sure if its relevant to the problem, but maybe the definition of "data" object is an issue (need to remove one bracket)?

data = read("./)pro.gro") == should be ==> data = read("./pro.gro")

[LijieZhong]

Oh,I made a simple mistake. I think the error occurred when I copy this code, but there's actually no brackets in the local version.

[LijieZhong]

I uploaded a structure file.
pro.zip