Pygenesig, a framework to generate and validate tissue-specific gene signatures.¶
Gene signatures are sets of genes derived from gene expression data, which identify a certain tissue, cell type, pathway, etc. Pygenesig provides a framework to create and validate such signatures. The package is easily extensible to add new methods for signature creation and testing.
Getting started¶
Please refer to the documentation, in particular the sections
If you want to implement own methods for creating and testing signatures, please take a look at the
Developer’s guide and the
Installation¶
You need to have Python 3.7 or newer installed on your system. Some methods for creating or testing signatures additionally require R. If you don’t have Python installed, we recommend installing Miniforge.
There are several alternative options to install pygenesig:
Install pygenesig in a self-contained conda environment:
This is the most reliable option to make both R and Python work. Make sure you have the `conda-forge` and the `bioconda` channels set-up with the correct priorities as described in the Bioconda documentation.
# use `mamba` instead of `conda` for more speed mamba create -n pygenesig python=3.8 pip bioconductor-edger bioconductor-bioqc conda activate pygenesig pip install pygenesig
Install pygenesig via pip and R packages manually
pip install pygenesig
Then, in R:
install.packages("BiocManager") BiocManager::install(c("edgeR", "BioQC"))
Usually, if
R
is in yourPATH
, rpy2 automatically detects your R installation. If you get an error message while importingpygensig
, try setting theR_HOME
environment variable before importing pygenesig:import os os.environ["R_HOME"] = "/usr/lib/R" import pygenesig
Install the latest development version from GitHub:
pip install git+https://github.com/grst/pygenesig.git@master
You’ll need to separately install R packages as described above.
Release notes¶
See the release section.
Contact¶
Please use the issue tracker.