1) Installing NGS_RNA

We first have to load EasyBuild, this can be done with this command

module load EasyBuild

The NGS_RNA pipeline has a lot of dependencies, these are handled by easybuild when the --robot command is executed (all the dependencies can be found here). Since we have also our own repo we have to give the path to that also. There can be multiple paths to easybuild configs, just separate them by colon.

Note: the order in which you give the paths are important! To original easybuild path can be left empty (just a colon is enough)

eb NGS_RNA-3.2.4.eb --robot --robot-paths=${pathToMYeasybuild}/easybuild-easyconfigs/easybuild/easyconfigs/:

Note: some software cannot be downloaded automagically due to for example licensing or technical issues and the build will fail initially. In these cases you will have to download manually and copy the sources to ${HPC_ENV_PREFIX}/sources/[a-z]/NameOfTheSoftwarePackage/ This is the case for example for Java. Therefore:

scp jdk-7u80-linux-x64.tar.gz [email protected]:${root}/apps/sources/j/Java/
scp jdk-8u45-linux-x64.tar.gz [email protected]:${root}/apps/sources/j/Java/

but also tools as GATK, Tabix and snpEff should be download manually:

scp GATK-3.5.tar.gz [email protected]: ${root}/apps/sources/g/GATK/
scp tabix.0.18.6.tar.gz [email protected]: ${root}/apps/sources/t/tabix

2) Installing the necessary resources (reference genome, dbSNP etc)

Logout and login again. Run the script RNA_resources to install the required resources, you can download the scripts here

sh NGS_RNA-resources.sh

This script will download parts of the 2.8 bundle from the GATK server

Note: Sometimes the GATK ftp server can be down/instable, try it a couple of times

3) Creating workdir structure

sh makestructure.sh