Screening and replication using the same data set: testing strategies for family-based studies in which all probands...

Area of Science:

• Genetics
• Statistical genetics
• Genomics

Background:

• Family-based designs are crucial for genome-wide association studies (GWAS).
• Existing methods like Family-Based Association Tests/Transmission Disequilibrium Tests (FBATs/TDTs) have limitations in power and applicability.

Purpose of the Study:

• To introduce a powerful two-stage testing strategy for GWAS in parent-offspring trio data where all offspring are affected.
• To enhance the detection of genetic variants associated with complex diseases.

Main Methods:

• A two-stage approach: Stage 1 estimates genetic effect size independently of FBATs/TDTs and computes conditional power.
• A weighted Bonferroni procedure assigns significance levels based on power estimates.
• Stage 2 tests SNPs using FBATs/TDTs at individually adjusted significance levels.

Main Results:

• Simulation studies show substantial power increases compared to standard methods (e.g., FBATs/TDTs with Bonferroni correction).
• The strategy is effective even when the true genetic model is unknown.
• Application to a childhood asthma GWAS identified two genome-wide significant markers missed by standard methods.

Conclusions:

• The proposed two-stage strategy offers a significant advancement in GWAS power for family-based studies.
• This method improves the detection of disease-associated genetic markers, with practical implications for complex disease research.