A quantile-based method for association mapping of quantitative phenotypes: an application to rheumatoid arthritis...

Area of Science:

• Genetics
• Statistical Genetics
• Rheumatoid Arthritis Research

Background:

• Traditional genetic association studies for quantitative traits often use Analysis of Variance (ANOVA).
• ANOVA's reliance on specific statistical assumptions can lead to false-positive results in genetic association.
• Quantitative traits like anti-CCP and RF-IgM are crucial biomarkers for rheumatoid arthritis.

Purpose of the Study:

• To explore and evaluate model-free statistical methods as alternatives to ANOVA for genetic association analysis.
• To assess the performance of correlation-based methods in identifying genetic associations for rheumatoid arthritis-related quantitative traits.
• To compare the validation rates of ANOVA versus model-free correlation methods using permutation testing.

Main Methods:

• Employed correlation statistics between allele frequencies and quantitative trait quantile values as a model-free approach.
• Conducted genome-wide association scans on anti-cyclic citrullinated peptide (anti-CCP) and rheumatoid factor-immunoglobulin M (RF-IgM) quantitative traits.
• Utilized data from the Genetic Analysis Workshop 16 (GAW 16) for analysis.

Main Results:

• Significant genetic associations were detected for both anti-CCP and RF-IgM on Chromosome 6, excluding the known rheumatoid arthritis susceptibility locus in the Human Leukocyte Antigen (HLA) region.
• A substantial portion of significant findings from ANOVA did not validate under permutation testing.
• A higher proportion of significant findings from the correlation statistic were validated using permutations, indicating greater robustness.

Conclusions:

• Model-free correlation methods provide a robust alternative to ANOVA for genetic association studies of quantitative traits.
• The developed correlation approach demonstrates improved reliability in identifying true genetic associations compared to traditional ANOVA.
• Further investigation into these model-free methods can enhance the accuracy of genetic discovery in complex diseases like rheumatoid arthritis.