Citation | Installation | Example | Usage | Command line | API doc | Interface classes | Website | Acknowledgment

Try nglview online:

An IPython/Jupyter widget to interactively view molecular structures and trajectories. Utilizes the embeddable NGL Viewer for rendering. Support for showing data from the file-system, RCSB PDB, simpletraj and from objects of analysis libraries mdtraj, pytraj, mdanalysis, ParmEd, rdkit, ase, HTMD, biopython, cctbx, pyrosetta, schrodinger's Structure

Should work with Python 3. If you experience problems, please file an issue.

Ask question about usage? Please post here

Table of contents

• Installation
• Example
• Showcase from users
• Usage
• Contributing
• Command line
• API doc
• Interface classes
• Changelog
• FAQ
• Website
• Acknowledgment
• Cite
• License

Installation

Released version

• Available on conda-forge channel

  conda install nglview -c conda-forge
  # might need: jupyter-nbextension enable nglview --py --sys-prefix
  
  # if you already installed nglview, you can `upgrade`
  conda upgrade nglview --force
  # might need: jupyter-nbextension enable nglview --py --sys-prefix

• Available on PyPI

   pip install nglview
   jupyter-nbextension enable nglview --py --sys-prefix

To use in Jupyter Lab you need to install appropriate extension:

jupyter labextension install  nglview-js-widgets

But please read this sample script first: https://github.com/arose/nglview/blob/master/devtools/nglview-jupyterlab.sh

Notes

If you are using notebook v5.0, you need to increase the iopub_data_rate_limit to visualize big structure (e.g: solvated system)

jupyter notebook --NotebookApp.iopub_data_rate_limit=10000000

Development version

Requirement: ipywidgets >= 7.0, notebook >= 4.2

The development version can be installed directly from github:

notebook user

    git clone https://github.com/arose/nglview
    cd nglview
    python setup.py install

    # if you edit files in ./js folder, make sure to rebuild the code
    cd js
    npm install

    # probably need to activate widgetsnbextension
    # python -m ipykernel install --sys-prefix
    # jupyter nbextension enable --py --sys-prefix widgetsnbextension
    # jupyter nbextension enable --py --sys-prefix nglview

    # tested with ipywidgets 5.2.2, notebook 4.2.1

Example

• Notebooks: please see our Jupyter notebook examples
• Simple demo for trajectory (take time to load): biomembrane

Showcase from users

Please check user examples. Feel free to contribute.

Usage

Open a notebook

jupyter notebook

and issue

import nglview
view = nglview.show_pdbid("3pqr")  # load "3pqr" from RCSB PDB and display viewer widget
view

A number of convenience functions are available to quickly display data from the file-system, RCSB PDB, simpletraj and from objects of analysis libraries mdtraj, pytraj, mdanalysis, ParmEd, rdkit, HTMD, biopython.

Function	Description
show_file(path)	Shows any NGL supported file formats (pdb, gro, mol2, sdf, dx, ..) in path
show_pdbid(pdbid)	Shows pdbid fetched from RCSB PDB
show_simpletraj(struc_path, traj_path)	Shows structure & trajectory loaded with simpletraj
show_mdtraj(traj)	Shows MDTraj trajectory traj
show_pytraj(traj)	Shows PyTraj trajectory traj
show_parmed(structure)	Shows ParmEd structure
show_mdanalysis(univ)	Shows MDAnalysis Universe or AtomGroup univ
show_rdkit(mol)	Shows rdkit rdkit.Chem.rdchem.Mol
show_ase(atoms)	Shows ase Atoms
show_asetraj(traj)	Shows ase trajectory traj
show_pymatgen(struct)	Shows pymatgen Structure
show_htmd(mol)	Shows HTMD Molecules
show_biopython(mol)	Shows Biopython structural entities
show_iotbx(mol)	Shows cctbx's iotbx structure
show_rosetta(pose)	Shows pyrosetta's Pose
show_iodata(obj)	Shows iodata's IOData
show_psi4(obj)	Shows psi4's Molecule
show_qcelemental	Shows QCelementary's Molecule
show_openbabel	Shows openbabel's OMol
show_prody	Shows prody's Ensemble or AtomGroup

API

Representations

view.add_representation('cartoon', selection='protein')

# or shorter
view.add_cartoon(selection="protein")
view.add_surface(selection="protein", opacity=0.3)

# specify color
view.add_cartoon(selection="protein", color='blue')

# specify residue
view.add_licorice('ALA, GLU')

# clear representations
view.clear_representations()

# update parameters for ALL cartoons of component 0 (default)
view.update_cartoon(opacity=0.4, component=0)

# remove ALL cartoons of component 0 (default)
view.remove_cartoon(opacity=0.4, component=0)

# Not using default representation
view = nv.show_file('your.pdb', default=False)
view.center()
view.add_rope()

Representations can also be changed by overwriting the representations property of the widget instance view. The available type and params are described in the NGL Viewer documentation.

view.representations = [
    {"type": "cartoon", "params": {
        "sele": "protein", "color": "residueindex"
    }},
    {"type": "ball+stick", "params": {
        "sele": "hetero"
    }}
]

The widget constructor also accepts a representation argument:

initial_repr = [
    {"type": "cartoon", "params": {
        "sele": "protein", "color": "sstruc"
    }}
]

view = nglview.NGLWidget(struc, representation=initial_repr)
view

Properties

# set the frame number
view.frame = 100

# parameters for the NGL stage object
view.stage.set_parameters(**{
    # "percentages, "dist" is distance too camera in Angstrom
    "clipNear": 0, "clipFar": 100, "clipDist": 10,
    # percentages, start of fog and where on full effect
    "fogNear": 0, "fogFar": 100,
    # background color
    "backgroundColor": "black",
})

# note: NGLView accepts both origin camel NGL keywords (e.g. "clipNear")
# and snake keywords (e.g "clip_near")

# parameters to control the `delay` between snapshots
# change `step` to play forward (positive value) or backward (negative value)
# note: experimental code
view.player.parameters = dict(delay=0.04, step=-1)

# update camera type
view.camera = 'orthographic'

# change background color
view.background = 'black'

Trajectory

# adding new trajectory
view.add_trajectory(traj)
# traj could be a `pytraj.Trajectory`, `mdtraj.Trajectory`, `MDAnalysis.Universe`,
# `parmed.Structure`, `htmd.Molecule` or derived class of `nglview.Trajectory`

# change representation
view[0].add_cartoon(...) # equal to view.add_cartoon(component=0)
view[1].add_licorice(...) # equal to view.add_licorice(component=1)

Add extra component

# Density volumes (MRC/MAP/CCP4, DX/DXBIN, CUBE)
# Or adding derived class of `nglview.Structure`
view.add_component('my.ccp4')

# add component from url
view.add_component('rcsb://1tsu.pdb')
# NOTE: Trajectory is a special case of component.

Mouse

# coot mouse style (https://en.wikipedia.org/wiki/Coot_(software))
view.stage.set_parameters(mouse_preset='coot')

Interaction controls

• Mouse
• Keyboard

Movie making

Require: moviepy (pip install moviepy)

from nglview.contrib.movie import MovieMaker
movie = MovieMaker(view, output='my.gif', in_memory=True)
movie.make()

Embed widget

embed

API doc

• Latest version
• All releases versions
• Development version

Command line

# open a notebook and import nglview
nglview

# Require installing pytraj (PR for other backends is welcome)
# open notebook, load `my.pdb` to pytraj's trajectory then display `view`
nglview my.pdb

# load density data
nglview my.ccp4

# open notebook, create trajectory with given topology `my.parm7` and trajecotry file `traj.nc`,
# then display `view`
nglview my.parm7 -c traj.nc

# load all trajectories with filename ending with 'nc'
# make sure to use quote " "
nglview my.parm7 -c "*.nc"

# open notebook, copy content from `myscript.py`
nglview myscript.py

# create a remote notebook
# just follow its instruction
nglview my.pdb --remote
nglview my.parm7 -c traj.nc --remote
nglview mynotebook.ipynb --remote

# demo (don't need pytraj)
nglview demo

# specify web browser
nglview my.pdb --browser=google-chrome

FAQ

Q&A

Website

• http://nglviewer.github.io/nglview/latest
• http://nglviewer.github.io/nglview/dev

Talks

Talks about NGL and nglview

Contributing

Join us here

Projects integrating NGLView

(Feel free to make a PR to add/remove your project here. Thanks.)

• AMBER - A package of programs for molecular dynamics simulations of proteins and nucleic acids
• mbuild - A hierarchical, component based molecule builder
• deepchem - Deep-learning models for Drug Discovery and Quantum Chemistry
• htmd - High throughput molecular dynamics simulations
• Moleidoscope - Molecular kaleidoscope
• ssbio - Tools for enabling structural systems biology
• hublib - hublib is a Python library for the HUBzero science gateway platform.
• molPX: ipython API to visualize MD-trajectories along projected trajectories
• nanoribbon
• ase: Atomic Simulation Environment
• pida: Software for analyzing multiple protein-protein interaction docking solutions,
• pytim
• MobleyLab/drug-computing Educational materials for, and related to, UC Irvine's Drug Discovery Computing Techniques course.
• pyiron: an integrated development environment for implementing, testing, and running simulations in computational materials science.
• BioSimSpace: An interoperable framework for biomolecular simulation
• pyrod: PyRod - Tracing water molecules in molecular dynamics simulations
• kugupu: kugupu - a molecular network generator to study charge transport pathways in amorphous materials
• pnab: proto-Nucleic Acid Builder
• opencadd: A Python library for structural cheminformatics
• teachopencadd: TeachOpenCADD: a teaching platform for computer-aided drug design (CADD) using open source packages and data
• cando: Computer Aided Nanostructure Design and Optimization
• query.libretexts.org: query.libretexts.org

Acknowledgment

• Funding: Hai Nguyen is supported by NIH Grant GM103297, "The Center for HIV RNA Studies" (2015 to 02-2017).
• Many thanks to nglview contributors
• dunovank/jupyter-themes: for oceans16 theme
• base64-arraybuffer
• ipywidgets

Cite

If you would like to acknowledge our work, feel free to cite:

Hai Nguyen, David A Case, Alexander S Rose; NGLview - Interactive molecular graphics for Jupyter notebooks, Bioinformatics, , btx789, https://doi.org/10.1093/bioinformatics/btx789

License

Generally MIT, see the LICENSE file for details.