Ravages

Ravages applies multinomial regression–based extensions of burden tests to detect associations between rare genetic variants and multicategory phenotypes.

Key Features:

• Multinomial Regression Model: Employs multinomial regression to model associations across multicategory phenotypes and assess heterogeneity among case subgroups.
• Burden Test Extensions: Extends burden tests, including CAST and WSS, to accommodate rare-variant aggregation in multicategory outcome analyses.
• Type I Error and Power Evaluation: Evaluates type I error rates and statistical power under simulated scenarios to assess test performance.
• Application to Genetic Heterogeneity: Enables subgroup-stratified analyses (e.g., age-of-onset groups) and demonstrated improved burden-test performance in Moyamoya disease.

Scientific Applications:

• Genetic association studies: Testing associations between rare variants and complex traits with multicategory phenotypes.
• Analysis of genetic heterogeneity: Detecting subgroup-specific variant effects across case subgroups such as different age-of-onset groups.
• Rare variant aggregation: Aggregating rare variants for association testing using extended burden statistics like CAST and WSS.
• Disease-specific analysis: Applied to Moyamoya disease as an example of stratified rare-variant association analysis.

Methodology:

Integrates burden tests (CAST, WSS) with multinomial regression models and uses simulation-based evaluation of type I error and statistical power, with subgroup stratification by phenotype categories such as age-of-onset.