rBahadur

rBahadur: High-dimensional binary data generation under structured genetic correlations

Assortative Mating Simulation Tool models non-random mating and generates high-dimensional binary random variates using the Bahadur order-2 approximation of the multivariate Bernoulli distribution to reproduce complex genetic correlation structures.

Key Features:

• Assortative Mating Simulation: Simulates non-random mating where individuals with similar phenotypes or genotypes preferentially mate, enabling modeling of population structure effects.
• Bahadur Order-2 Approximation: Generates high-dimensional binary random variates using the Bahadur order-2 approximation of the multivariate Bernoulli distribution to preserve higher-order correlation structure.
• Correlation Structures: Supports general and diagonal-plus-low-rank (DPLR) correlation matrices for flexible modeling of genetic dependencies.
• rb_dplr Algorithm: Implements the rb_dplr algorithm for computationally efficient large-scale synthetic dataset generation.

Scientific Applications:

• Population Genetics Studies: Evaluates genetic consequences of assortative mating on population structure and diversity.
• Evolutionary Biology Research: Assesses effects of non-random mating on speciation and adaptation processes.
• Genetic Architecture Analysis: Investigates how alternative correlation structures influence complex trait and disease architecture.

Methodology:

Generates high-dimensional binary random variables via the Bahadur order-2 approximation to the multivariate Bernoulli distribution, parameterized with general or diagonal-plus-low-rank correlation structures. Correlation patterns induced by assortative mating are encoded in the specified covariance structure, and the rb_dplr algorithm enables efficient sampling for large-scale genotype and phenotype simulations.