A new multimarker test for family-based association studies.

Area of Science:

• Genetics
• Statistical Genetics
• Bioinformatics

Background:

• Family-based studies are crucial for identifying genetic associations.
• Candidate gene approaches require efficient statistical testing strategies.
• Evaluating the impact of various factors on statistical power is essential.

Purpose of the Study:

• To develop and evaluate a novel multimarker test for family-based candidate gene studies.
• To compare the performance of different testing strategies, including tag SNP selection and family designs.
• To identify the optimal strategy for maximizing statistical power in genetic association studies.

Main Methods:

• Simulations under diverse genetic models were used to assess testing strategy performance.
• Factors evaluated included statistical tests, tag SNP methods, number of tag SNPs, and family designs.
• An Analysis of Variance (ANOVA) model summarized the effects of these factors on statistical power.

Main Results:

• Tag SNP methods, gene characteristics, and family designs had minimal impact on the optimal strategy.
• The familywise error rate (FWER) controlling procedure and the new multimarker test demonstrated the highest power.
• Both FWER and the multimarker test showed increased power with more tag SNPs and were invariant to family designs.

Conclusions:

• The optimal strategy for family-based candidate gene studies involves using the FWER or the new multimarker test.
• Selecting 6-10 tag SNPs using any considered tag SNP method is recommended for maximizing power.
• The findings were validated through an application to Alzheimer's disease data.