Hydra - a flexible RNASeq pipeline.

Features

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

Running the pipeline

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.

Note

If you're using a module system such as on the shared computing cluster (SCC) at Boston University you can just load a preinstalled version:

module purge
module load anaconda2/4.3.0

2. Set Environment and Build Paths:

export CONDA_ENVS_PATH=<WORKING_DIR>/conda-envs
export CONDA_BLD_PATH=<WORKING_DIR>/conda-bld

3. Create Dev Environment

Use conda to install a basic developement environment:

conda env create \
  montilab/dev_env \
  -p ./dev_env

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)

4. 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:

require(devtools)
install_git("http://github.com/nachocab/clickme.git")

Running the pipeline

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 .

Additional documentation

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