ptc-cmc / cybercamp

Rapid overview of molecular simulation

Cybercamp

The cybercamp is rapid overview of molecular simulation and the underlying tools.

Overview

This repository is designed to provide users the resources to quickly get up to speed with molecular simulations. This will not only include information needed to better understand simulations, but it will also provide specific information for workflow management, "sandboxed" development environments, python specific scientific packages, etc.

Getting Started

Prerequisites

• miniconda

  • A python package mangement tool, useful for creating clean, reproducible development environments

• Command line familiarity

  • Most of the interaction with simulation programs involves some knowledge of the GNU/Linux, or Unix command line interface.

  • A useful guided tutorial for the command line can be found here.

• Visualization tools

  • Humans are much more adept at visually inspecting data when it is presented properly. We are not great at reading a large data file and developing any meaningful hypotheses/conclusions from a list of numbers. We are much better at looking at plot of the data, or rendering the trajectory of a molecule evolving in space and time.
  • To do this we require molecular visualization tools and data visualization tools
  • For molecular visualization, a common choice is VMD
  • A common pythonic choice for data visualization is Matplotlib and can easily be installed via anaconda

Installing

To get a working development environment that uses the MoSDeF toolkit, HOOMD-Blue for simulation, Matplotlib for data visualization, and signac and signac-flow for workflow management, we will use conda we installed earlier.

Update conda

conda update conda

Add the conda-forge channels for conda to search through when installing packages.

conda config --add channels conda-forge

Create a new python environment named cybercamp2022 that includes many of the packages needed to build systems, run simulations, and analyze the data.

conda create -n cybercamp2022 mbuild foyer hoomd matplotlib py3Dmol gsd  signac signac-flow fresnel openbabel jupyter jupyterlab

Activate the environment

conda activate cybercamp2022

Content

Computational Basics

• Unix Basics
• Conda and Git
• Python: Basics
• Python: Working with Numpy
• Python: Working with Matplotlib
• Python: Introduction to Pandas
• Python: Pandas-Seaborn

The following links are to Software Carpentry courses on the use of the unix/linux shell, Python, and plotting using matplotlib.

• Introduction to the Unix shell: http://swcarpentry.github.io/shell-novice/
• Introduction to Python: http://swcarpentry.github.io/python-novice-inflammation/
• Plotting and Programming in Python: http://swcarpentry.github.io/python-novice-gapminder/

Molecular Modeling and Simulation

• Introduction to Simulation
  • General overview of simulation
  • Discussion of common systems studied with molecular simulation
  • Interactive demo/exercises of a simple simulation of the motion of a ball
• Introduction to Molecular Dynamics
  • Overview of models and methods used in molecular dynamics simulation
• Anatomy of a Script File
  • Discussion of common input parameters needed by molecular dynamics simulation codes
  • Example script files for HOOMD-Blue and LAMMPS
  • Hands on demo/exercises performing simple LJ simulation using HOOMD-Blue
• Neighborlists and Dangerous Builds
  • Discussion of a basic neighborlist
  • Hands on example of changing parameters and their impact
• Predicting Phase Behavior of a LJ system using MD
  • Hands on exercises performing MD simulations at different statepoints
  • Basic analysis of simulation results

Other resources

• Introduction to version control with git: http://swcarpentry.github.io/git-novice/
• Using Jupyter notebooks:
  • http://jupyter.readthedocs.io/en/latest/
  • https://jupyter-notebook-beginner-guide.readthedocs.io/en/latest/
  • https://www.datacamp.com/community/tutorials/tutorial-jupyter-notebook
• Brief guides on GitHub and git workflows: https://guides.github.com/
• Introduction to version control with git: http://swcarpentry.github.io/git-novice/
• Full list of software carpentry lessons: https://software-carpentry.org/lessons/
• Bevan Lab GROMACS Tutorials: http://www.mdtutorials.com/gmx/

Primary Authors

• Christopher R. Iacovella
• Justin Gilmer
• Andrew Z. Summers

License

This project is licensed under the MIT License - see the LICENSE.md file for details

cybercamp's Issues

Installation Instructions are in two different places

The main README.md page has left over instructions on how to set up a Conda environment. However, this is also part of the setting_up_you_computer directory that we point to to include all of the steps for setting things up. We will want to consolidate this into one place.

Link to gromacs tutorials is broken

The bevan lab vt link is no longer relevant - it has moved to http://www.mdtutorials.com/gmx/

I can't fork this repo

https://media.giphy.com/media/3o6gb3kkXfLvdKEZs4/giphy.gif

Timestep Optimization won't work with VMD

Current versions of VMD (1.9.4a) which are necessary as previous versions are 32 bit programs, cannot open gsd trajectory files. As such, VMD visualization in this notebook must be done using a different method. One recommendation would be to use a gsd plugin for VMD, https://github.com/mphowardlab/gsd-vmd, or to output a dcd trajectory (hoomd.dump.dcd()) and load in a topology as well for easy visualization.