Public | Automated Build

Last pushed: a year ago
Short Description
Systems biology working environment
Full Description

sysbio-docker

Serves the purpose of having some systems biology simulation and coversion tools in one place. \
Features:

  1. libSBML
  2. libroadrunner for Python
  3. antimony
  4. Local scripts for simulation and plotting.

Installation

This repository is linked to Docker hub via an automated build process.
Having all of these tools in Docker format enables easier management of the tools necessary across different Linux machines.

From Github

You can clone this repo and build the Docker locally:

git clone https://github.com/abulovic/sysbio-docker.git
cd sysbio-docker
docker build -t abulovic/sysbio-docker .

This will create a Docker image named sysbio which you can find if you invoke

docker images

From Docker Hub

You can clone the docker image directly from Docker hub:

docker run -ti abulovic/sysbio-docker

After this, you will have a abulovic/sysbio-docker image listed under docker images.

Running the image

The way this docker image is being used now is to:

  • Run interactive shell through Docker with all tools installed
  • Share certain local folders with the Docker image
  • Simulate models and export the data to the shared folders

You can run the Docker image by running the provided run-docker.sh script.

How it works

So, the idea is that you put the models you want to simulate in the local ./models directory in the SBML (XML) format. You need to put the models there prior to running the docker:

./run-docker

Let us presume the name of your model is called model.xml. The simulator needs the model file, but it can also accept configuration in json format and the output directory to which the simulation results will be stored:

./code/simulator [PATH TO MODEL FILE] --cfg [PATH TO JSON CONFIG] --odir [OUTPUT DIRECTORY]

In your case:

./code/simulator ./models/model.xml --odir ./output

The configuration is not necessary, but useful, as I'll explain in a second. If successful, it will give you the output of a format:

Results stored to:  ./output//ModelName/2016-06-06T20:37:32.778831

You can browse the folder and see that the results of the simulation are stored in the ; delimited format in a textual file.
You can use this file to generate plots in an interactive IPython notebook using Plotly-offline.
Run the following command with the output folder from the previous script:

./code/create-notebook ./output//ModelName/2016-06-06T20:37:32.778831

You can now check the content of that folder - you should have a plotter.ipynb

Docker Pull Command
Owner
abulovic
Source Repository