MDAnalysis was accepted for Google Season of Docs 2019. We
are proud to be one of 50 participating open source
organizations, which span all reaches of open
source and includes the Wikimedia Foundation, NumPy and SciPy, the
CERN High Energy Software Foundation, Drupal, Arduino, LLVM Compiler
Infrastructure, and many more outstanding projects that really have
been moving the open source community forward.
We would love to have a technical writer work with us to improve our
documentation. If you want to work with scientists that write code to
do science at the molecular scale then please contact us on the
developer list.
More about Projects and MDAnalysis
Our recent post on our GSoD application
explains more of the background and summarizes our project ideas.
MDAnalysis is a project written by scientists for scientists and is
currently used for cutting edge research around the world.
A writer with MDAnalysis will learn about the
technical aspects of creating docs for open source code (such as
documentation generation with Sphinx, continuous integration-based
workflows and distributed development with git via our GitHub
source code repository) and also about the science that MDAnalysis
enables. Our developers have published many scientific publications
that used our library or described innovative new methods that were
implemented in MDAnalysis. A writer will interact with these
accomplished scientists and learn about the science: How
molecules are simulated (e.g., biomolecules such as proteins and DNA or
novel materials) and how scientists analyze such simulations in order
to better understand the world at the microscopic scale.
The video shows the result of a molecular dynamics simulation in which
a sodium ion (magenta sphere) together with water molecules
(red/white “L” shapes) bind to a protein (shown in
cartoon representation as blue and green
ribbons), a process that is important in maintaining the health of
cells in the body. The MDAnalysis package enables us to write code
which can perform calculations on these results,
here allowing us to quantify the interaction
between the ion and the protein and generate fundamental understanding
of important processes in living beings.
Contact
If you have any questions or if you want to work with us please write to us on the
developer list.
MDAnalysis is applying for Google Season of Docs 2019. We would
love to have a technical writer work with us to improve our
documentation.
In this program, Google is sponsoring a professional technical writer
to work with an open source project. The open source project gets to
work with a professional in the area of documentation — something
that almost all projects struggle with and want to improve — and
the writer gets experience working with an open
source project, e.g., being embedded in the developer community and
using tools such as automatic doc generation and continuous
integration services.
Because MDAnalysis is a project written by scientists for
scientists, a writer with MDAnalysis would not only learn about the
technical aspects of creating docs for open source code (such as
documentation generation with Sphinx, continuous integration-based
workflows and distributed development with git via our GitHub source code
repository) but also about the science that MDAnalysis
enables. Our developers have published many scientific publications
that used our library or described innovative new methods that were
implemented in MDAnalysis. A writer would interact with these
accomplished scientists and learn about the science: How
molecules (especially biomolecules such as proteins and DNA) are
simulated and how scientists analyze such simulations in order to
better understand the world at the microscopic scale.
What is MDAnalysis?
MDAnalysis is a Python library for the analysis of computer
simulations of many-body systems at the molecular scale.
MDAnalysis is also a community of hundreds of users and developers. As
a community, we adhere to our Code of Conduct.
The goal of MDAnalysis is to make it easy for users to analyze
data that are produced by simulations (primarily molecular
dynamics simulations) that run on some of the largest
super-computers in the world. We accomplish this goal by providing a
toolkit of programming building blocks that provide an abstract
interface to the simulation data that lends itself to interactive data
exploration and rapid prototyping but is also a robust foundational
library that can form the basis for new computational
tools.
Scientists in our field use statistical mechanics to analyze the
simulation data and make predictions that connect theory and
experiments. For example, the image shows simulated data (points with
error bars) compared to a theoretical model (solid line). The
background shows the protein that was simulated. MDAnalysis was used
for the analysis.
Project Ideas
Our goal for GSoD is to make it easy for new users to analyze
their data. We want to accomplish this goal by
providing a quick introduction to the essentials of using
MDAnalysis through (1) overview documentation and (2) tutorials
that address common areas of interest of users;
connecting the introductory docs to the API docs so that users can
easily learn and explore by themselves.
We have listed several possible projects to work on on our
wiki. A summary table is shown below:
project
name
effort
description
1
User story based documentation
75%
Create documentation (starting from existing MDAnalysis docs) addressing common well-known use cases of the library. The structure should be at a higher level than the existing module-level default documentation and similar to the structure used for libraries such as scikit-learn and pytorch.
2
Introduction to analyzing Molecular Dynamics trajectories with MDAnalysis
25%
Create documentation centered around the 2016 SciPy talk by Beckstein (video, slides) with notebooks illustrating the fundamentals of molecular dynamics and how MDAnalysis facilitates analyzing such simulations.
3
Quick Start Guide
25%
Create a guide for getting started with MDAnalysis within a Jupyter notebook in less than 10 minutes. Includes installation, data loading, and sample real-world use case. Base on MDAnalysis Docs: Overview and MDAnalysis Tutorial.
4
Beginner, Intermediate, and Advanced Tutorials
50%–100%
Reorganize the existing documentation into Beginner, Intermediate, and Advanced material that build upon each other. The tutorials should progress from (beginner) basic trajectory analysis to (intermediate) working with topology information to (advanced) System building (see the selection of topics). The material should include code references in GitHub, static or live Jupyter notebooks, and illustrations to facilitate learning and understanding.
The column effort is a rough estimate of what percentage of one
GSoD could be spent on this project. Technical writers can combine a
lower effort project (as an introductory project) with one of the
high-effort projects. Projects with effort ranges are modular in that
we can work on different independent sub-components and thus tailor
the effort.
If you want to learn more, have a look at our projects page,
which contains more details, or contact us and ask us.
Contact
If you have any questions or if you want to work with us please write to us on the
developer list.
While our documentation is mostly focused on using MDAnalysis for exploratory
analysis it is equally well suited to build your own analysis library on top of
it. Below is a list of all projects we know about that use MDAnalysis.
Visualization tools
nglview: nglview is a tool to visualize
trajectories in jupyter notebooks.
mda-pymol:
MDAnalysis has been embedded into PyMOL to read many different MD formats
directly
Analysis tools
pydiffusion: Analyze the
rotational diffusion of your molecules.
pytim: Pytim is a package based on
MDAnalysis for the identification and analysis of surface molecules in
configuration files or in trajectories from molecular dynamics simulations.
pycontact: Analysis of
non-covalent interactions in MD trajectories.
pyPcazip:
A PCA-based toolkit for compression and analysis of molecular simulation data
RotamerConvolveMD:
Analysis of molecular dynamics trajectories or conformational ensembles in
terms of spin-label distances as probed in double electron-electron resonance
(DEER) experiments.
PBxplore: PBxplore is a suite of
tools dedicated to Protein Block (PB) analysis.
cgheliparm: Scripts used to analyze
dsDNA structures from Martini MD simulations.
