Code repository for "The individual and global impact of copy number variants on complex human traits"
https://doi.org/10.1101/2021.08.10.21261839
Contact: Chiara Auwerx (chiara.auwerx -at- unil.ch) or Zoltan Kutalik (zoltan.kutalik -at- unil.ch).
Burden_analysis: Contains the scripts to calculate each individual's autosomal CNV burden (CNV_burden) and determine which individual carries a CNV overlapping CNV regions (CNVRs) identified to be significantly associated to a trait through CNV-GWAS (CNVR_covariates). Resulting data is used to perform CNV burden analyses on continuous traits (continuous; both with and without correction for modifier CNVs), as well as life history traits (life_history).
CNV_calling: Contains the pipeline for raw BAF and LRR data processing, CNV calling with PennCNV, quality score (QS) calculation, and transformation of QSs to the probe level.
CNV_inheritance: Contains the scripts to calculate the fraction of CNVs individuals in our main analysis share with their UKBB siblings or random individuals.
EstBB: Contains the scripts to perform simulations aiming at establishing the power of our replication study in the Estonian Biobank.
GWAS: Contains script for sample selection (01_samples), probe selection (02_probes), covariate (03_covariates) and phenotype (04_phenotypes) extraction and correction, GWAS with PLINK (05_gwas), identification of independent signals through stepwise conditional analysis (06_SCA), CNVR definition (07_CNVR) and annotation (08_ANNOVAR), and locus zoom on specific candidate CNVRs (10_locus_zoom).
CNV frequency and summary statistics available at:
http://dx.doi.org/10.17632/z54dc3b6jz.1
Alternatively, summary statistics can be downloaded from the GWAS Catalog under the study accession numbers: GCST90027274-GCST90027444.