EBglmnet: a comprehensive R package for sparse generalized linear regression models

Bioinformatics. 2016 Mar 15;btw143. doi: 10.1093/bioinformatics/btw143. Online ahead of print.

Abstract

EBglmnet is an R package implementing empirical Bayesian method with both lasso (EBlasso) and elastic net (EBEN) priors for generalized linear models. In our previous studies, both EBlasso and EBEN outperformed other state-of-the-art methods such as lasso and elastic net in inferring sparse genotype and phenotype associations, in which the number of covariates is typically much larger than the sample size. While high density genetic markers can be easily obtained nowadays in genetics and population analysis thanks to the advancements in molecular high throughput technologies, EBglmnet will be a very useful tool for statistical modeling in this area.

Availability and implementation: EBglmnet package is freely available from the R archive CRAN (http://cran.r-project.org/).

Contact: a.huang1@umiami.edu.