Biomolecular Simulation: OpenMM and MDAnalysis: Setup

Highlights

• Obtain the Lesson Materials
• Using Google Colab for Python
  • Highlights
  • Using Python Packages
• Installing Python (Anaconda) Locally
  • Installing Anaconda Packages
  • Installing Into an Environment
• AMBERTools for Structure Preparation
  • Conda Installation of Tools Binaries
  • Full Source Code

Obtain the Lesson Materials

All files have been made available on GitHub. Click the green button labeled Clone or download and select Download ZIP. Open the ZIP folder and move the files wherever you would like to work from.

Using Google Colab for Python

Any of the code contained herein can be run using Google Colab. You can install Python packages using !pip install in a code cell. Colab notebooks are interactive Python notebook, much like Jupyter notebooks.

Any files that you need to use within a Colab notebook can be on your Google Drive account. For files the Drive, click the folder icon on the left of the code/text cells, and select “Mount Drive.” If you don’t want to link your Google Drive, you can select “Upload” instead.

Highlights

• Text-based cells are written in Markdown
• Cells can be evaluated by Shift+Enter
• Lines can be easily commented with Cntl+/ ( Cmd+/ on Mac)

Using Python Packages

A number of commonly used packages are already available for import in the Google Colab environment, like matplotlib, numpy, and pandas. Additional packages can be installed through !pip install XXX or !apt-get install XXX where XXX is the name of the package to be installed. Additional information about this can be found here.

Installing Python (Anaconda) Locally

1. Go to the Anaconda Download page.
2. Download the installer for your operating system. (You likely use 64-bit.) Make sure you get the Python 3 version.
3. Double click the installer and follow the instructions. All the default options are fine for MacOS and Linux. For Windows, be sure to select Make Anaconda the default Python (the rest of the defaults are fine).

Installing Anaconda Packages

Packages can be installed through conda install XXX where XXX is the package name. In certain cases, you have to specify a channel.

For OpenMM:

conda install -c omnia -c conda-forge openmm

For MDAnalysis:

conda config --add channels conda-forge
conda install mdanalysis

Installing Into an Environment

OpenMM and MDAnalysis have a number of dependencies, and sometimes they conflict with what packages are currently installed. If you’ve previously installed Anaconda, then it is advised that you create a conda environment for these tutorials. If you’ve never used the environments feature before, you’ll need to run conda init <SHELL> to properly reset your ~/.bashrc (ex. conda init bash).

conda create --name openmm2020 python=3.6
conda activate openmm2020
conda install -c omnia -c conda-forge openmm
conda install mdanalysis -c conda-forge

The name section can be anything, you’re not limited to openmm2020, and the python version can be whatever the newest version of 3.X is installed.

In the future, you use the environment by running conda activate openmm2020. To leave the environment, run conda deactivate.

AMBERTools for Structure Preparation

Conda Installation of Tools Binaries

AMBERTools has added a Python-based installation for the tools binaries. It should work for Linux and MacOS, but is not a true substitute for the full AMBERTools code. It is fine for the purposes of this tutorial.

conda install -c conda-forge ambertools

Full Source Code

The full source-code can be downloaded from the AMBER website. This is the version that should be linked with an AMBER distribution.