molPX
molPX copied to clipboard
IPython API to visualize MD-trajectories along projected trajectories inside a Jupyter notebook
################################################### Welcome to molPX: The Molecular Projection Explorer ################################################### |DOI| |travis_build| |appveyor_build| |coverage| |docs_build|
The Molecular Projection Explorer, molPX, is a python module that provides interactive visualization of
projected coordinates of molecular dynamics (MD) trajectories inside a Jupyter notebook <http://jupyter.org/>_.
molPX is based on the incredibly useful nglview IPython/Jupyter widget <https://github.com/arose/nglview>.
Other libraries heavily used are mdtraj <http://mdtraj.org/> and PyEMMA <http://www.emma-project.org/latest/>.
At the moment, there is also an sklearn <http://scikit-learn.org/stable/index.html> dependency that might disappear in the future.
.. image:: ../images/output.gif :align: center
At the moment the API consists of two subpackages:
- :doc:
molpx.visualize </index_visualize> - :doc:
molpx.generate <index_generate>
TL;DR: see molPX in action through the
- :doc:
Example Jupyter Notebooks </index_notebooks>
Find more about the people behind molPX here:
- :doc:
About & YouTube Introduction </about>
Download and Install
If you can't wait to play around with molPX, and you have the Anaconda scientifc python distribution <https://www.continuum.io/downloads>_ (which we strongly recommend), the easiest way to get molPX is to issue
the conda command <https://conda.io/docs/intro.html>_:
conda install molpx -c conda-forge
and jump to the Quick Start section of this document. Otherwise, check out our more exhaustive
- :doc:
Installation Guide </INSTALL>
Quick Start
Start an IPython console
>>> ipython
Import molpx and let the example notebooks guide you
>>> import molpx
>>> molpx.example_notebooks()
Voilà: you should be looking at a list of jupyter notebooks explaining the basic functionality of molPX
Documentation
You can find the latest documentation online here <https://molpx.readthedocs.io/>_
You can build a local copy of the html documentation by navigating to the molPX installation
directory and issuing:
>>> cd doc
>>> make html
This will generate molPX/docs/build/html/index.html with the html documentation. If you are missing some of
the requirements for the documentation , issue:
>>> pip install -r ./source/doc_requirements.txt
If you don't know where molPX is installed, you can find out this way:
>>> ipython
>>> import molpx
>>> molpx._molpxdir()
The output of the last command is one subdirectory of molPX's installation directory, so just copy it and issue:
>>> cd the-output-of-the-molpx._molpxdir-command
>>> cd ..
and you are there !
Warnings
molPX is currently under heavy development and the API might change rapidly. Stay tuned.
Data Privacy Statement
When you import this Python package, some of your metadata is sent to our servers. These are:
- molPX version
- Python version
- Operating System
- Hostname/ mac address of the accessing computer
- Time of retrieval
How to disable this feature easily:
Even before you use molPX for the first time:
- Create a hidden folder
.molpxin your home folder - Create a file
conf_molpx.pyinside of.molpxwith the following line:report_status = False - Restart your ipython/jupyter sessions
Hints:
-
This is most easily realized from terminal by issuing:
>>> mkdir ~/.molpx >>> echo "report_status = False" >> ~/.molpx/conf_molpx.py -
You can check your report status anytime by typing this line in a (i)python terminal
>>> import molpx >>> molpx._report_status() -
If you don't know where your home folder is (for whatever reason), you can find it out by typing in a (i)python terminal
>>> import os >>> os.path.expanduser('~/.molpx')
.. |DOI| image:: https://zenodo.org/badge/76460348.svg :target: https://zenodo.org/badge/latestdoi/76460348 :height: 20 :alt: DOI
.. |travis_build| image:: https://travis-ci.org/markovmodel/molPX.svg?branch=master :height: 10 :target: https://travis-ci.org/markovmodel/molPX :alt: Travis build status
.. |appveyor_build| image:: https://ci.appveyor.com/api/projects/status/396ej39s3fewhwy9/branch/master?svg=true :height: 10 :target: https://ci.appveyor.com/project/gph82/molpx :alt: Appveyor build status
.. |coverage| image:: https://codecov.io/gh/markovmodel/molPX/branch/master/graph/badge.svg :height: 20 :target: https://codecov.io/gh/markovmodel/molPX :alt: Codecov
.. |docs_build| image:: https://readthedocs.org/projects/molpx/badge/?version=latest :alt: Documentation Status :height: 20 :target: http://molpx.readthedocs.io/en/latest/?badge=latest