FastBiCmrMLM

FastBiCmrMLM implements a compressed variance component mixed logistic model for large genomic case-control genome-wide association studies of binary traits.

Key Features:

• Compressed variance components: The model compresses four variance components into two to reduce computational complexity while retaining all effects and polygenic backgrounds.
• Three integrated computational algorithms: Algorithms are tailored to sample size, computational speed, and reduced memory requirements for large-scale data analysis.
• FastBiCmrMLM-Hap extension: Linkage disequilibrium markers are replaced by bin-based haplotypes to mine additional genes.
• Detection of challenging variants: The method identifies dominant variants, small allele substitution effects (alpha ≈ 0), and rare variants.
• Ultra-rare variant detection: FastBiCmrMLM detects variants with frequencies as small as 0.03% in UK Biobank-scale datasets.

Scientific Applications:

• Large-scale binary trait GWAS: FastBiCmrMLM analyzes large genomic case-control datasets to decipher the genetic architecture of binary traits.
• Re-analysis of existing disease datasets: Re-analysis of seven diseases in the WTCCC datasets identified 36 novel variants and 29 candidate genes with functional and TWAS evidence.
• Haplotype-based association mapping: FastBiCmrMLM-Hap enables additional gene discovery through bin-based haplotype analysis of linkage disequilibrium regions.

Methodology:

FastBiCmrMLM includes all effects and polygenic backgrounds in a mixed logistic model for binary traits and compresses four variance components into two. The compressed model combines three computational algorithms tailored to sample size, computational speed, and reduced memory requirements. FastBiCmrMLM-Hap replaces linkage disequilibrium markers with bin-based haplotypes.