• features:
  • user-defined fastq-splitting for faster computation
  • opional UMI-filtering (fgbio pipeline)
• Outline: map(BWA) > dedup(UMI/standard) > indelRealign(gatk) > baseRecalibrate(gatk) [> UMI consensus call]
• cluster config for SLURM and SGE
• used as base for WESSomVar

• copy files from setup/local into your local setup
• copy stuff into your own .bash*
• adjust for OS dialect
• produce your .ssh folder following the BIH guideline enter the BIH server directly to computation node

$ ssh bihcluster 

• provide password $ qrsh | srun --pty bash -i
• ready
• copy .bash_profile .bashrc .condarc to your $HOME

• install miniconda:

$ wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh

• set to your path

$ bash Miniconda3-latest-Linux-x86_64.sh -b -f -p $HOME/work/miniconda

• add $HOME/work/miniconda/bin to PATH:

$ 'export PATH=$PATH:$HOME/work/miniconda/bin' >> ~/.bashrc

• refresh bash:

$ source ~/.bashrc

• activate conda base environment:

$ . activate base # or conda activate base

• create or move to project folder for cloning pipeline repository into
• git clone this repo into that folder:

$ git clone https://github.com/Mar111tiN/somVarGATK.git

• use env-file (hwes.yml) to create new environment (rename in first line to <your_env>)

$ conda env create -f setup/WES.yml
$ . activate <your_env>
$ . deactivate

• env is stored centrally in ../miniconda/envs/<your_env>
• executables are stored/symlinked in ../miniconda/envs/<your_env>/bin

!! while in that environment!!!:

• add missing conda channels to ~.condarc file
• install what you want using conda install ...
• [ place symlinks to your files into .../miniconda/envs/<your_env> ] ( = $CONDA_PREFIX/envs/<your_env>)
• add shell-scripts to PATH:

export $PATH:<folder>/scripts

• ready to go

• a sample config_blueprint.yaml is provided to guide you through the configuration of the tools
• make sure all reference and annotation files are existing (snakemake will tell you if not)