The following instructions cover Docker installation of both pipelines: BiG-SCAPE and CORASON. BiG-SCAPE can also be installed manually or within a virtual environment. Docker is a container platform provider available for multiple operative systems such as Mac OS X, Windows 10 and Linux (Fedora, Ubuntu, Centos and Debian). The installation procedure uses `curl`. See curl installation if you do not have it installed.

Docker engine installation

Install docker engine

BiG-SCAPE and CORASON can run on docker. If you have docker engine installed, please skip this step. This is a Linux minimal docker installation guide, if you do not use Linux or you are looking for a detailed tutorial on Linux/Windows/Mac Docker engine installation please consult Docker: Getting Started.

$ curl -fsSL | sh

Then type:
$ sudo usermod -aG docker your-user

Important step:
Log out from your session (restart your machine) and get back in into your user session before the next step. You may need to restart your computer and not just log out from your session in order to changes to take effect.Test your docker engine with the command:
$ docker run hello-world

BiG-SCAPE installation

At the moment, there are two methods available to run BiG-SCAPE:
- Using the pre-built Docker image, a slightly larger download but zero-fuss install on any system that can run Docker.
- Installing all the dependencies (using a virtual environment is recommended).

Installation using Docker

$ mkdir ~/bin # not required if you already have that
$ curl -q > ~/bin/run_bigscape
$ chmod a+x ~/bin/run_bigscape
$ ~/bin/run_bigscape

If BiG-SCAPE is correctly installed the terminal displays the help menu after some seconds:

Manual installation


* Python (Python 3.x. compatibility with Python 2 is not guaranteed)
* The HMMER suite, version 3.2 (or any version compatible with Pfam)
* The (processed) Pfam database. For this, download the latest `Pfam-A.hmm.gz`
file from the Pfam website , uncompress it and process it using the `hmmpress` command.
* Biopython
* Numpy
* scipy
* scikit-learn
* NetworkX

Installing dependencies using Conda

Although each library can be installed individually, the use of a virtual environment is highly recommended. Here is a quick guide for BiG-SCAPE installation using Miniconda (downloading the Pfam library manually is still necessary):
* Install Miniconda . We recommend that you install the Python 3 version as default for all new conda environments. You will need to login again for the changes to go into effect. You can download Miniconda as follows:


* Create a new environment. You can choose Python 3 using `python=3` at the end of the next command even if you installed the Python 2 version of Miniconda (see additional documentation of the conda environment here .

$ conda create --name bigscape

* Activate new environment:

$ source activate bigscape

* Install packages:

$ conda install numpy scipy scikit-learn
$ conda install -c bioconda hmmer biopython fasttree
$ conda install -c anaconda networkx

Once the environment is ready (or the packages installed), download or clone the code from the repository and run as

$ python [parameters]

(see here for an overview of the options) .

Corason installation

Docker installation

$ mkdir ~/bin # not required if you already have that
$ curl -q > ~/bin/run_corason
$ chmod a+x ~/bin/run_corason
$ ~/bin/run_corason

If CORASON is correctly installed, the terminal displays the help menu after some seconds: