ConSReg 1.1.7

condition-specific regulation

ConSReg 1.1.4

Condition-specific regulations

• ConSReg 1.1.4
• Getting Started
  • 1. Installation
    • 1.1 Required packages
      • 1.1.1 Python
      • 1.1.2 R
    • 1.2 Easy installation by Anaconda (recommended)
    • 1.3 Manual installation (Skip this section if 1.2 is successful)
    • 1.3.1 R installation
      • install R
      • install R packages
    • 1.3.2 Python installation
  • 2. Sample datasets
  • 3. Analysis
  • 4. Publication

Table of contents generated with markdown-toc

Getting Started

1. Installation

1.1 Required packages

1.1.1 Python

• python = 3.6
• numpy == 1.16.2
• scipy == 1.1.0
• pandas == 0.21.1
• joblib >= 0.11
• rpy2==2.8.6
• networkx >= 2
• sklearn >= 0.19.1
• intervaltree == 2.1.0

1.1.2 R

• ChIPSeeker == 1.16.1
• CoReg == 1.0.1
• gglasso == 1.4
• RRF == 1.9
• R >= 3.5.1

1.2 Easy installation by Anaconda (recommended)

Since ConSReg is dependent on both Python and R packages, we recommend installing ConSReg by Anaconda to easily set up the running environment. You may retrive Anaconda from here and install the version corresponding to your OS.
Once Anaconda is installed in your OS, run the following commands to create an new environment and install ConSReg and all its dependencies into the new environment:

conda create -y -n consreg python=3.6 # The new environment name is 'consreg'. You may use other name instead.
conda activate consreg
conda install -y -c bioconda --no-channel-priority bioconductor-chipseeker
conda install -y --no-channel-priority r-base r-essentials
conda install -y --no-channel-priority -c conda-forge r-gglasso r-rrf r-devtools
pip install ConSReg

Then ConSReg environment can be activated by conda activate consreg and disabled by conda deactivate

1.3 Manual installation (Skip this section if 1.2 is successful)

1.3.1 R installation

install R

If R is not already installed， you may follow these steps to build R from source code. Otherwise, you may skip this section and start from 1.2.2

First, disable any conda environment, if there is an active one.

conda deactivate

Download R source code from CRAN (https://cran.r-project.org/). You may use any version you like. It is recommended to use R version > 3.0.0. This ensures that rpy2 works correctly with R.

# Download R 3.6.1
wget https://cran.r-project.org/src/base/R-3/R-3.6.1.tar.gz

Decompress the downloaded file

tar -zvxf R-3.6.1

In the decompressed folder, configure R by:

./configure prefix=path_to_install_R --enable-R-shlib

--prefix= specifies a writeable directory to install R into. --enable-R-shlib flag was added to build R shared libraries.

In the decompressed folder, compile R

make

Install R into the specified directory:

make install

Add a line to ~/.bashrc to tell the OS where to look for R

export PATH=path_to_R_bin_directory:$PATH

Add the following line to ~/.bashrc. This is for telling rpy2 where to look for dynamic libraries.

export LD_LIBRARY_PATH=/home/alexsong/R/3.6.1/lib64/R/lib:$LD_LIBRARY_PATH

Apply the changes to environment variables PATH and LD_LIBRARY_PATH:

source ~/.bashrc

install R packages

ConSReg requires several R packages: ChIPseeker, CoReg, gglasso and RRF.

It is recommended to deactivate any conda environment when installing R packages, as it may add the environment-specific path which may fail the installation. If any conda environment is active, you may deactivate it by:

conda deactivate

To install ChIPSeeker from bioconductor, type the following commands in R (for R 3.6 or higher version):

if (!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("ChIPseeker")

For older version of R, type the following commands in R:

source("https://bioconductor.org/biocLite.R")
biocLite("ChIPseeker")

Please refer to the instructions described here for more details.

To install CoReg pakcage from GitHub, type the following commands in R environment:

install.packages("devtools")
library(devtools)
install_github("LiLabAtVT/CoReg")

Please refer to the GitHub page of CoReg project for more details: link

To install gglasso package from CRAN, type the following commands in R environment:

install.pacakges("gglasso")

Please refer to the link here for more details.

To install RRF package from CRAN, type the following commands in R environment:

install.pacakges("RRF")

Please refer to the link here for more details.

1.3.2 Python installation

ConSReg can be installed by pip:

pip install ConSReg

Sometime rpy2 may throw out error message when imported in Python. This problem may arise because rpy2 was built with the R version that is different from the one it is linked to when imported in Python. To fix this, you may remove rpy2 package then reinstall it with 'no-cache-dir' flag:

pip install ConSReg --no-cache-dir

Alternatively, you may want to install ConSReg in development mode to be able to edit the package by yourself. To do so, simply git clone this repository and then under the directory that contains setup.py, type in:

pip install -e .

2. Sample datasets

Sample datasets can be found in data folder.

3. Analysis

We provide code for analyzing the sample datasets in two jupyter notebooks located in the root folder of this project: bulk_analysis.ipynb (for bulk RNA-seq data) and single_cell_analysis.ipynb (for single cell RNA-seq data).

4. Publication

Please cite the followint paper if you use ConSReg in your research:
Qi Song, Jiyoung Lee, Shamima Akter, Ruth Grene, Song Li. "Prediction of condition-specific regulatory genes using machine learning." Nucleic acids research 48.11 (2020): e62-e62.