Extending rare-variant testing strategies: analysis of noncoding sequence and imputed genotypes.

Area of Science:

• Genetics
• Bioinformatics
• Statistical genomics

Background:

• Next Generation Sequencing (NGS) enables studying rare genetic variations in heritable traits.
• Existing single-marker tests lack power for rare variant detection.
• Pooling methods improve power but often require high-coverage sequencing.

Purpose of the Study:

• To introduce and evaluate the Cumulative Minor-Allele Test (CMAT) as a powerful and efficient pooling statistic.
• To extend CMAT for use with probabilistic genotypes and low-coverage sequencing data.
• To demonstrate CMAT's utility in analyzing imputed rare variants in existing GWAS datasets.

Main Methods:

• Simulated population genetic models to assess CMAT performance against other pooling methods.
• Extension of CMAT to handle probabilistic genotypes and low-coverage sequencing data.
• Application of CMAT to imputed rare variants using external imputation templates (1000 Genomes Project).

Main Results:

• CMAT achieves power comparable to existing methods across various study designs.
• Augmenting sequence data with imputed samples increases the power of rare-variant studies.
• CMAT can effectively analyze rare variants imputed into existing GWAS datasets, as demonstrated on the GAIN psoriasis dataset.

Conclusions:

• CMAT is a powerful and computationally efficient method for rare variant association studies.
• The extended CMAT is practical for low-coverage sequencing and imputation data.
• Imputation strategies significantly enhance the power of rare variant discovery in large-scale genetic studies.