Adjusted sequence kernel association test for rare variants controlling for cryptic and family relatedness

Area of Science:

• Genetics and Genomics
• Statistical Genetics
• Bioinformatics

Background:

• Advances in sequencing technology enable identification of rare variants linked to complex traits.
• Existing rare variant association methods often assume independence, limiting their application in family-based studies.
• Family-based designs can enrich datasets for familial rare disease variants but require specialized analytical approaches.

Purpose of the Study:

• To introduce a novel framework for association testing of rare variants within family-based designs.
• To adapt existing methods to effectively control for familial relatedness in genetic association studies.
• To develop a statistically robust and computationally efficient method for analyzing rare variants in families.

Main Methods:

• Developed an adjusted sequence kernel association test (ASKAT) framework.
• Integrated the SKAT approach with factored spectrally transformed linear mixed models (FaST-LMMs) to account for family structure.
• Utilized a linear mixed model (LMM) incorporating genome-wide identity by descent (IBD) and restricted maximum likelihood (REML) for variance component estimation.

Main Results:

• Simulation studies demonstrated that ASKAT provides robust control of type I error across varying heritability levels.
• The method exhibited good statistical power for detecting rare variant associations in family-based cohorts.
• ASKAT, leveraging FaST-LMM, demonstrated computational efficiency, enabling analysis of large-scale genomic data with hundreds of thousands of markers.

Conclusions:

• ASKAT offers a powerful and efficient solution for rare variant association testing in family-based studies.
• The method effectively addresses the challenge of population structure and relatedness in genetic analyses.
• The ASKAT methodology is applicable to large datasets, as illustrated by its application to UK Twins Consortium data.