giorkala/gnh_flagship_recessive

Scripts used for the recessive association analysis, within the G&H Flagship paper

Rare Recessive Burden Analysis

Scripts used for the recessive association analysis, within the G&H Flagship paper. Prior to this, we phased our WES dataset using a Snakemake pipeline developed here.

This repository contains:

1. get_biallelic_carriers.sh to identify biallelic carriers, given phased genotypes. Also, see here for a more involved script that starts from the phased genotypes and creates VCF files ready for association.
2. get_ch_homz_counts.sh generates a tally with the numbers of compound-het and homozygous genotypes.
3. concat_vcfs.sh concatenates chromosome-based VCFs to one, for a given consequence, and creates a BGEN for more efficient association testing.
4. run_regenie_{burden/sv/perm}.sh are scripts to invoke REGENIE for gene-burden testing, variant-level, or permutation (ie FDR analysis), respectively. Each script has several modules, e.g. step1 (when needed) or step2, for quant or binary traits, which are selected based on the arguments, e.g. bash run_regenie.sh step2qt pLOF will run step2 for quant traits and test for the pLOF burdens.
5. dominance_deviation_{linear/logistic}.R: R scripts to test for dominance deviation, for a candidate gene-trait pair. This requires a table with ADD/DOM/REC encodes, per gene, which can be obtained with dominance_deviation_prep.py.

Georgios Kalantzis, gk18@sanger.ac.uk
August 2024