Installation Guide
MutEnricher is now available as a Docker image on Docker Hub. Please see the Running with Docker page for instructions and examples.
The remainder of this page explains how to install MutEnricher on your system from the source code.
This software has been explicitly tested with Python 2.7 (versions 2.7.12 and greater) and Python 3.7 (versions 3.7.3) on Red Hat >=6, Ubuntu 16 LTS, and macOS Sierra. Compatibility with Python versions < 2.7 is likely possible, though untested. The library argparse, which is part of the standard Python library for versions >= 2.7, must be installed (in addition to the modules described below) if attempting use with versions < 2.7. Other packages and/or updates may also be necessary when running with older versions.
If not already configured on you system, we recommend installing Python with the Anaconda distribution for your system. This should cleanly install the Python environment, related tools, and the majority of the required packages necessary for running this code. Depending on your operating system configuration, additional programs may need to be installed manually, including OpenSSL, libssl, and libcrypto.
Beyond the basic standard Python library, several easily accessible Python packages must be installed (if not already present) for successful execution (some version restrictions noted):
The easiest way to obtain these additional packages is to use the Python package manager pip. Python versions >= 2.7.9 should include pip by default; otherwise, this package can be installed separately on your system (https://pip.pypa.io/en/stable/installing/). If Python was installed with Anaconda, pip should also be available and only pysam and cyvcf2 should need to be installed as extra packages.
If the cyvcf2 package does not install cleanly with pip on your system, you may need to ensure the OpenSSL, libssl, and libcrypto are appropriately installed/available on your system. Please see the cyvcf2 installation page for further information and instructions.
Install the required python packages (listed above).
This package utilizes “cythonized” code to enhance speed for some numerical functions. The user needs to compile these functions with his or her own version of Python/Cython to successfully run the programs:
- From the package install directory, enter the
math_funcssub-directory:cd math_funcs - Run the following command to cythonize the code:
python setup.py build_ext --inplace
These steps will produce the file math_funcs.so in the math_funcs directory.
The package should now be successfully installed on your system. The main package driver script is mutEnricher.py. To see the main help page, use python mutEnricher.py -h. To see the detailed help pages for the two analysis types, use python mutEnricher.py <command> -h, where <command> is either coding or noncoding.
We provide additional utilities (in the utilities sub-directory) that can be used to help generate feature covariate files for use with the programs. Currently, this directory includes the Python scripts:
- get_gene_covariates.py
- get_region_covariates.py
These scripts use the same input gene/region files (GTF or BED files, more on these later) to calculate sequence (and other) covariates for the genes/regions considered in the analyses. Further usage instructions for these optional scripts are described in more detail in the quickstart guide and tutorial.