This workshop source repository contains the build recipe for a docker container derived from the CCPBioSim JupyterHub image. This container adds the necessary software packages and notebook content to form a deployable course container.
This workshop introduces the application of MDTraj to the analysis of protein simulation data
It is in two parts:
-
INTRODUCTION An introduction to the use of MDTraj for the analysis of simulations. The data comes from the associated 'protein simulation with amber and openmm' workshop.
-
EQUILIBRATION An introduction to a key issue in MD simulation: testing for equilibration, sampling, and convergence. The data here comes from a long simulation of a different protein, chignolin.
We have a vision of MD simulations as consisting of an initial equilibration phase, during which system variables relax to "realistic" values, followed by the "production" phase from which valid estimates of features or behaviours of interest can be extracted. So how do we decide when an MD simulation is long enough that it is equilibrated, and data collection can begin?
This workshop looks at one of the most common methods for assessing equilibration: the calculation of RMSDs. It may be a common approach, but is it a good one? Maybe not.
A basic knowledge of Python and MD simulation methods
This training course is deployed on the CCPBioSim website via our cloud infrastructure, however you can deploy on your own machine with docker.
Pull the container from our repository::
docker pull ghcr.io/ccpbiosim/protein-analysis-workshop:latest
In our containers we are using the JupyterHub default port 8888, so you should forward this port when deploying locally::
docker run -p 8888:8888 ghcr.io/ccpbiosim/protein-analysis-workshop:latest
Workshop Content Authors:
- Charlie Laughton
Please direct all questions and feedback to Charlie Laughton