Setup resources
ENEO heavily depends on public genetic databases for germline probability estimation plus other fairly common resources daily needed for bioinformatics. In order to make the download and configuration of resources for the workflow as smooth as possible, a python configuration script is available inside setup/download_res.py. As the workflow uses files with different annotations, downloading resources is not enough, as they may need a chromosome naming conversion: thus some common used tools like bcftools are required. To run the automatic setup, first create a conda environment using the setup_env.yml file located inside the setup folder.
Then activate the conda environment, and launch the configuration script as follows, replacing the resources folder with .
Note
The genome assembly in use is the GRCh38. We're not planning to backport it to older assembly.
This script will download different resources: - genome, transcriptome and GTF from Gencode - 1000G, ExAC from the GATK resource bundle - dbSNPs ALFA - REDI portal - VEP cache (v105)
ENEO uses also COSMIC, to further annotate known somatic variants with the relevant COSMIC id. According to the nature of your entity (Commercial or Non-Commercial), please download the cosmic VCF file following the guidelines on the website. After downloading, add the PATH of the downloaded VCF into the configuration file (located under config/config_main.yaml)
Note
If you're not interested in using COSMIC and are not interested in adding this notation, you had to edit the file in workflow/utils/vcfanno.toml. In details, you had to remove the lines 31-35 and the lines 43-51
What if I already got some of them?
The configuration script works by controlling the existence of the files whose path is written inside the main configuration file conf_main.yaml, located in the config folder. If any of those files are already in your machine, just edit the configuration file adding the right absolute path. The script will check for its presence without re-downloading it.
Failure
The annotation for files must be concordant throughout the pipeline, to avoid any error. Be sure to add resources manually only if you're sure of their concordance.