ParseCNV2

ParseCNV2 performs comprehensive detection, curation, and association analysis of copy number variations (CNVs) in genome-wide association studies (GWAS).

Key Features:

• Multi-format input: Natively accepts sequencing-derived variant files in VCF format and SNP array CNV calls in PennCNV format.
• Scalability and performance: Optimized for speed and efficiency to handle cohorts exceeding 100,000 samples and biobank-scale datasets (e.g., UK Biobank n > 450,000; CAG Biobank n > 350,000) genotyped at >0.5 million probes without dividing data across compute clusters.
• Hypothesis-free GWAS support: Performs genome-wide CNV analyses without pre-selecting candidate variants to consider the full spectrum of CNVs.
• Pre- and post-association CNV curation: Implements rigorous CNV quality-control and curation steps before and after association testing.
• Clinical CNV prioritization and QC: Provides clinical CNV prioritization and quality-control features to support clinical research applications.
• Covariate adjustment: Supports adjustment for covariates in association analyses to reduce confounding.
• GWAS catalog integration: Formalizes CNV association methodology for inclusion alongside SNP associations in GWAS catalogs.

Scientific Applications:

• Genome-wide CNV association studies: Detection and association testing of CNVs across the genome in GWAS datasets.
• Disease association studies: Identifying CNVs associated with disease phenotypes through association testing.
• Large-scale biobank analysis: Analysis of CNVs in biobank-scale cohorts such as UK Biobank and CAG Biobank.
• Clinical CNV interpretation: Prioritization and quality-control of CNVs for clinical research and clinical-grade analyses.
• Hypothesis-free discovery: Unbiased discovery of CNV associations without predefining candidate regions.

Methodology:

Accepts VCF and PennCNV inputs; performs pre- and post-association CNV curation and quality control; conducts hypothesis-free genome-wide CNV association testing with covariate adjustment and produces CNV association outputs formalized for GWAS catalogs, optimized for large-cohort processing.