ld-block-removal

Software requirements

To run this pipeline, you must have nextflow and singularity installed. This can be done using the conda environment env.yaml. First conda must be installed, which can be done following the instructions here: https://docs.anaconda.com/miniconda/miniconda-install/. Once installed, the environment can be created as follows:

conda env create --name env.yaml

Once installed, activate the environment in order to run the pipeline by:

conda activate nextflow

Setup

Configuration

Input data and parameters are specified in a run-specific configuration file created by the user. An example is found in test/test.config, and can include the following parameters:

• COHORT [ default = "UKB" ] : Genomic cohort you are working with. Current supported cohorts are UKB and GENOMICC. These are configured to format the chromosome values as they are represented in the BGEN files for a given cohort. If UKB or GENOMICC are not specified, the default will be tot ake the integer value for each chromosome, with no prefix.
• INPUT_SNPS [ required ] : Path to a CSV file, where each line represents a SNP-of-interest. Columns must be RSID, CHR, POS. RSID must be in the format as it appears in the BGEN files.
• BGEN_FILES [ required ] : Path to BGEN files for the cohort-of-interest. Must be split by chromosome and specified as so: "/path/to/bgen/files/test_cohort_chr*.{bgen,sample,bgen.bgi}".
• BED_FILES [ required ] : Path to BED files to use for PCA after LD blocks have been removed. Must be split by chromosome and specified as so: "/path/to/bed/files/test_cohort_chr*.{bed,bim,fam}".
• TRAITS_DATASET [ required ] : see https://targene.github.io/targene-pipeline/stable/all_workflows_parameters/.
• RUN_TYPE [ default = "DISCOVERY" ] : Any non-test run.
• NB_PCS [ default = 50 ] : Number of PCs to compute usign FlashPCA2 after removing LD blocks.
• FLASHPCA_EXCLUSION_REGIONS [ default = assets/exclusion_regions_hg19.txt ] : Regions to exclude from BED files before PCA. hg38 also available in assets/exclusion_regions_hg38.txt.
• UKB_ENCODING_FILE [ default = "NO_UKB_ENCODING_FILE" ] : see https://targene.github.io/targene-pipeline/stable/all_workflows_parameters/.
• UKB_WITHDRAWAL_LIST [ default = "assets/NO_WITHDRAWAL_LIST" ] : see https://targene.github.io/targene-pipeline/stable/all_workflows_parameters/.
• UKB_CONFIG [ default = "assets/ukbconfig.yaml" ] : see https://targene.github.io/targene-pipeline/stable/all_workflows_parameters/.
• QC_FILE [ default = "assets/NO_QC_FILE" ] : see https://targene.github.io/targene-pipeline/stable/all_workflows_parameters/.
• MAF_THRESHOLD [ default = 0.01 ] : minor allele frequency threshold when filtering BED files during LD block removal.
• OUTDIR [ default = "results/" ] : Output directory for final results.

Run

Once your configuration file has been set up, you can run the pipeline at the command-line using nextflow.

This pipeline can be run with different profiles depending on what platform the pipeline is being run on. Current supported platforms/profiles are: eddie, ultra2, local.

nextflow run main.nf -c your_config.config -profile your_profile

Results

Results will be deposited in results/ by default, unless a different OUTDIR is specified in your configuration.