A New Method for Detecting Associations with Rare Copy-Number Variants

Area of Science:

• Genetics and Genomics
• Statistical Genetics
• Disease Etiology

Background:

• Copy number variants (CNVs) are significant contributors to diseases like cancer and psychiatric disorders.
• Existing methods for analyzing rare CNVs often fail due to their complexity and heterogeneity.
• Current CNV collapsing methods can be suboptimal as they ignore etiological heterogeneity.

Purpose of the Study:

• To introduce CCRET, a new random effects test designed for collapsing rare CNVs in disease association studies.
• To develop a method robust to etiological heterogeneity and applicable to multi-categorical variants.
• To provide a tool for collective evaluation of rare CNV effects on disease risk, accounting for confounders.

Main Methods:

• Developed CCRET, a random effects test for collapsing rare CNVs.
• Applied CCRET to variants measured on a multi-categorical scale.
• Conducted extensive simulations and analyzed large-scale schizophrenia datasets to evaluate CCRET's performance.
• Incorporated simultaneous correction for multiple confounders.

Main Results:

• CCRET demonstrates powerful and robust performance across various types of etiological heterogeneity.
• The method effectively models the collective effects of multiple CNV features.
• CCRET shows comparable or superior performance to existing methods, especially when heterogeneity is present.

Conclusions:

• CCRET offers a significant advancement in analyzing rare CNVs for disease association studies.
• The method's robustness to etiological heterogeneity makes it a valuable tool for genetic research.
• CCRET provides a more accurate assessment of CNV contributions to disease risk compared to existing burden tests.