KnockoffTrio

KnockoffTrio identifies putative causal genetic variants in father–mother–child trio designs by applying the knockoff framework to control the false discovery rate in family-based genetic association studies.

Key Features:

• False Discovery Rate Control: Controls the false discovery rate under arbitrary correlations among tests and is less conservative than family-wise error rate methods such as Bonferroni correction.
• Enhanced Power and Flexibility: Integrates with more powerful, potentially nonlinear models to increase power beyond standard family-based association tests.
• Robust Against Confounding: Provides protection against confounding from population stratification and distinguishes direct genetic effects from indirect effects.
• Prioritization of Causal Variants: Prioritizes putative causal variants over associations driven by linkage disequilibrium.

Scientific Applications:

• Large-scale family-based analyses: Applied to 14,200 trios from the Autism Genetic Resource Exchange (AGP), SPARK, and the Simons Simplex Collection (SSC) in autism spectrum disorder (ASD) cohorts, identifying associations not detected by conventional tests.
• Replication of Known Associations: Replicated ASD association signals with variants in MACROD2, NRXN1, PRKAR1B, CADM2, PCDH9, and DOCK4.
• Discovery of New Associations: Identified new associations at an FDR of 10% with variants in ARHGEF10, SLC28A1, ZNF589, and HINT1.

Methodology:

Creates synthetic knockoff copies of genetic features via the knockoff framework to control the FDR and accommodate complex correlations among tests in trio study designs.