Logica: A likelihood framework for cross-ancestry local genetic correlation estimation using summary statistics

Logica estimates local cross-ancestry genetic correlation from genome-wide association study summary statistics via a bivariate linear mixed model that accounts for ancestry-specific linkage-disequilibrium structure. Across simulations and 13 traits from multi-ancestry biobanks, Logica delivers greater accuracy, power, and false-discovery-rate control and yields valid genome-wide estimates in regions where existing methods fail.