R-Forge Logo

lrgpr: Low Rank Gaussian Process Regression

lrgpr is a high-peformance, user-friendly R interface for evaluating linear mixed models. This package is designed for interactive, exploratory analysis of large genome-wide assocation studies (GWAS) using linear mixed models to account for the confounding effects of kinship and population structure.

The package also provides simple interfaces for standard (i.e. fixed effects) linear and logistic regression models. It allows fitting millions of regression models on a desktop computer by using an efficient implementation, parallelization and out-of-core computing for datasets that are too large to fit in main memory.

Contact: gabriel [dot] hoffman [at] mssm.edu

Gabriel E. Hoffman. Updated September 24, 2019