Skip to content

Latest commit

 

History

History
94 lines (72 loc) · 4.33 KB

Installation.md

File metadata and controls

94 lines (72 loc) · 4.33 KB
Raw

Installation

Contents:

virtual environment

We recommend that RCK is installed in the isolated virtual environments. Virtual environments can be created via anaconda and python (w.r.t. RCK, which is written in Python).

To create a virtual environment (named rck-env) with anaconda, run the following command:

conda create --name rck-env python=3.7

To create a virtual environment (named rck-env) with python, run the following command:

python -m venv rck-env

If virtual environments are used (which, again, we recommend), we assume that the environment is activated.

conda (recommended)

Run the following conda command, to install RCK:

conda install -c aganezov rck

Installation via conda automatically takes care of Gurobi python bindings (refer to respective subsection), and everything shall work from this part (assuming that Gurobi is correctly installed and working).

pip

Run the following command, to install RCK:

pip install rck

WARNING: this installation does take care of python bindings for Gurobi. Please, refer to respective subsection on how that can be addressed.

source

First, download the source code. Example is shown below:

git clone https://github.com/aganezov/RCK.git

then run the following command from the RCK source folder:

python setup.py install

WARNING: this installation does take care of python bindings for Gurobi. Please, refer to respective subsection on how that can be addressed.

Gurobi

Gurobi solver can be obtained from the official web site and installation procedure is also described there. Gurobi requires a valid license to run. Licensing information is provided on the official website, and is available for free for individual academic users. More complicated setups with multi-user and cluster licenses are also available (and described on the official Gurobi website). Contact your university IT support for more information about any complication with Gurobi licensing and setup.

RCK expects that Gurobi is installed on the machine in question. RCK requires python bindings be installed (in the virtual environment, if you use it (which we recommend)). Refer to the next subsection for details on how this can be addressed.

Python bindings for Gurobi

RCK requires python bindings be installed (in the virtual environment, if you use it (which we recommend)). The following documentation of the Gurobi website explains how an installation of such bindings can be done.

Recommended way is via anaconda. Regardless of whether conda is used for virtual environment, os just in general, the following command will install Python Guorbi bindings:

conda install -c gurobi gurobi

If not using conda, one needs to go to the Gurobi installation dir and locate the setup.py file and run the following command:

python setup.py install

Note that this way is deprecated by Gurobi.

Executables

Installation of RCK adds several executables to your PATH (if using virtual environment, this executables will be accessible only when the environment is activated):

  • rck - main executable that runs the RCK inference algorithm
  • rck-adj-x2rck - conversion of SV prediction from several 3rd-party SV prediction tools (refer to respective docs section for more details)
  • rck-adj-process - various processing options for RCK formatted adjacencies
  • rck-scnt-x2rck - conversion of the clone- and allele-specific segment copy number predictions by 3rd-party tools (refer to respective docs section for more details)
  • rck-scnt-process - various processing options for RCK formatted segments, copy number, boundaries, etc.