Quickstart

LIBRA is a modular toolbox and hence easy to use. All outputs from functions and directories tree are generated behind scenes and required user interaction is low.

Follow this steps to a proper installation:

Install Python >=3.7.0.
Install R >=3.5.2.
Install sc_libra Python package.
OPTIONAL: Prepare the environment (Only for selecting a specific R version in case many are installed, otherwise avoid this step).
Importing sc_libra.

Installation

LIBRA is compatible with Python >=3.7, and depends on rpy2, NumPy, SciPy, Pandas and Keras. All required dependencies will be automatically installed when running:

$ pip install sc_libra

(Optional) Environment preparation

LIBRA makes use of rpy2 for running some specific R functions. In order to import properly this dependency is mandatory that Python knows where are the R libs for the specific R version used. This is a requirement of rpy2 and should be done, else sc_libra will raise an error when importing it (as it will import all dependencies required as it is imported) This can be done in two steps:

Run on console prior to run Python.

$ #Typical locations are:
$ # export LD_LIBRARY_PATH="/opt/R/3.5.2/lib64/R/lib:$LD_LIBRARY_PATH" (if local installation of R was done)
$ # export LD_LIBRARY_PATH="/usr/lib64/R/:$LD_LIBRARY_PATH" (if global installaltion of R was done)

$ export LD_LIBRARY_PATH="/YOUR_PATH_TO_R_LIBS_HERE:$LD_LIBRARY_PATH"

Open your Python environment and run:

$ import os
$ # Typical locations are:
$ # export os.environ['R_HOME'] = "/opt/R/3.5.2/lib64/R/" (if local installation of R was done)
$ # export os.environ['R_HOME'] = "/usr/lib64/R/" (if global installation of R was done)

$ os.environ['R_HOME'] = "/YOUR_POATH_TO_R_HERE"