This is a focused collection of makefiles and scripts to for analysis on Illumina sample data. While this probably can't be directly used by someone not in our lab, it might be a useful reference for others.

Set up

In order to set up these scripts, checkout the repository. There are some environment variables to set:

• STAMPIPES -- full path to the installation directory
• LIMS_API_URL -- URL to the LIMS API
• LIMS_API_TOKEN -- Auth to the LIMS API
• MODULELOAD -- Full path to the GNU modules init script
• HOTSPOT_DIR -- Path to HOTSPOT, defaults to ~/hotspot/hotspot-distr - download at [https://github.com/rthurman/hotspot]
• PYTHON3_ACTIVATE -- A virtualenv activation script for python3.

A python version that includes pip needs to be installed for proper python package version recording.

Modules

The current modules to load for working the pipeline are:

• bcl2fastq/1.8.4
• bcl2fastq2/2.15.0.4
• bedops/2.4.35-typical
• bedtools/2.25.0
• bowtie/1.0.0
• bwa/0.7.12
• coreutils/8.25
• cufflinks/2.2.1
• fastqc/0.11.5
• gcc/4.7.2
• git/2.3.3
• hotspot2/2.1
• htslib/1.6.0
• jdk/1.8.0_92
• kallisto/0.43.1
• `kentutil/302
• picard/2.8.1
• pigz/2.3.3
• pysam/0.9.0
• python/2.7.11
• python/3.5.1
• R/3.2.5
• RSEM/1.2.30
• samtools/1.3
• STAR/2.4.2a
• tophat/2.0.13
• numpy/1.11.0
• atlas-lapack/3.10.2
• scipy/1.0.0
• scikit-learn/0.18.1
• stringtie/1.3.4d

If $MODULELOAD is set, then the processing scripts should load the required modules automatically.