RetroFun-RVS: A Retrospective Family-Based Framework for Rare Variant Analysis Incorporating Functional Annotations

Area of Science:

• Genetics
• Bioinformatics
• Computational Biology

Background:

• Complex diseases often involve rare genetic variations in noncoding genome regions, posing interpretation challenges.
• Current rare variant association tests have limitations in power, design scope (case-control), and interpretation of regulatory mechanisms.

Purpose of the Study:

• To propose RetroFun-RVS, a novel retrospective family-based score test incorporating functional annotations for complex disease genetics.
• To enhance the power and interpretability of rare variant association tests using region-based functional data.

Main Methods:

• Developed RetroFun-RVS, a family-based score test aggregating genotypes to assess rare variant sharing among affected relatives.
• Integrated functional annotations, including 3D genome contacts and epigenetic data, into the analysis framework.
• Utilized extensive simulations and applied the method to a schizophrenia and bipolar disorder kindred study.

Main Results:

• RetroFun-RVS demonstrated superior power compared to existing region-wide and subregion methods, especially when integrating 3D genome contact networks.
• The method showed robustness to non-informative annotations and maintained power with variants spread across regions.
• A bootstrap procedure was recommended for small family sample sizes to mitigate Type I error inflation.

Conclusions:

• Integrating functional annotations, particularly networks with transcriptional impacts, significantly enhances rare variant tests for complex diseases.
• RetroFun-RVS offers a promising framework for identifying regulatory mechanisms underlying complex genetic disorders.
• The study highlights the utility of family-based designs and advanced functional data in genetic association studies.