Weighted functional linear regression models for gene-based association analysis.

Area of Science:

• Genetics
• Statistical genetics
• Bioinformatics

Background:

• Functional linear regression models are vital for gene-based association studies of complex traits.
• Existing methods often combine genetic variant information but can be limited by noise and error.
• Allele-specific weights have improved collapsing and kernel-based approaches.

Purpose of the Study:

• To introduce and evaluate allele-specific weights within functional linear regression models.
• To adapt these weighted models for both independent and family-based genetic samples.
• To assess the impact of weighting on the power and accuracy of gene-based association analyses.

Main Methods:

• Development of weighted functional linear regression models incorporating allele frequencies via beta distribution.
• Simulation studies using GAW17 genotypes to assess type I errors and power.
• Application of the weighted models to real-world blood pressure data from the ORCADES sample.

Main Results:

• Weighted models demonstrated type I errors consistent with declared values.
• Increasing weights for causal variants significantly enhanced the power of functional linear models.
• The new weighted method identified a novel association between diastolic blood pressure and the VMP1 gene (P = 8.18×10-6), outperforming unweighted and kernel-based models.
• Five of six known genes showed improved P values with weighted models in the ORCADES dataset.

Conclusions:

• Weighted functional linear regression models offer increased power for gene-based association analysis.
• The method is effective for both independent and family samples, improving the detection of genetic associations.
• The FREGAT package implements this novel weighted approach, providing a valuable tool for genetic research.