• html reporting
• automatic parallelization
• automatic resume function
• extensive quality control
• interactive principal component analysis

Pre-requisites

Hydra requires Java 1.6+. All other tools are wrapped in a Conda environment, see below.

1. Get Conda

conda is available as part of the miniconda package.

module purge
module load anaconda/2.2.0

2. Create Dev Environment

Use conda to install a basic developement environment:

conda create \
  -p ./dev_env \
  -c https://conda.anaconda.org/montilab \
  --yes \
  hydra

This will create a `./dev_env folder that stores all the files needed to run the pipeline. For more details, please refer to the Developer's Guide. (LINK)

3. Install all necessary R packages Activate the environment so you have access to the R version that the pipeline uses:

source activate ./dev_env

Start R:

R

Within R install all necessary packages:

source("http://bioconductor.org/biocLite.R")
biocLite(c("Biobase",'edgeR'),ask=F)
install.packages(c("devtools","knitr", "yaml", "rjson","codetools"), repos='http://cran.us.r-project.org')
devtools::install_github("nachocab/clickme")

Activate the environment (follow the instructions provided by conda following the environment's creation):

source activate ./dev_env

Then you can run the pipeline, while providing your parameter file.

Example:

hydra-rnaseq -p param.txt

A detailed description of the parameter file is provided here and an example is located here here .

A detailed user documentation is provided here. For advanced users that are contributing to the development such as adding additional modules a documentation is provided here