BGLR (A Statistical Package for Whole-Genome Regression)

Use Parametric and non-parametric regression with sets of high dimensional predictors. Whole-Genome Regression/Prediction.
Description

Use. The BGLR R-package implements a large collection Bayesian regression models, including various parametric regressions where regression coefficients are allowed to have different types of prior densities (flat, normal, scaled-t, double-exponential and various finite mixtures of the spike-slab family) and semi-parametric methods (Bayesian Reproducing kernel Hilbert spaces, RKHS). The software was originally developed as an extension of the BLR package (http://cran.r-project.org/web/packages/BLR/index.html ) and with a focus on genomic applications; however, the methods implemented are useful for many non-genomic applications as well. The response can be continuous (censored or not) or categorical (either binary, or ordinal).

Algorithms. The algorithm is based on a Gibbs Sampler with scalar updates and the implementation takes advantage of efficient compiled C and Fortran routines. The kernel of our software is written in R, but the computationally demanding steps are carried out using customized routines written in C and Fortran code. The implementation makes use of BLAS routines daxpy and ddot. The computational performance of the algorithm can be greatly improved if R is linked against a tuned BLAS implementation with multithread support, for example
OpenBLAS, ATLAS, Intel mkl, etc.

Data sets and ancillary functions. In addition to the main function (BGLR) the package comes with: (a) functions to read and write from the R-console *.ped and *.bed files, (b) two publicly available data sets comprising phenotypic, genomic and pedigree information and (c) various examples (see user’s manual below).
User’s Manual PDF file
Download package CRAN: http://cran.r-project.org/web/packages/BGLR/index.html
R-Forge: https://r-forge.r-project.org/R/?group_id=1525
Contact Gustavo de los Campos ( gcampos@uab.edu ) &
Paulino Pérez (perpdgo@gmail.com )
Related articles/
publications
  • de los Campos et al. (2013). Whole Genome Regression and Prediction Methods Applied to Plant and Animal Breeding. Genetics 193 (2): 327–345. doi:10.1534/genetics.112.143313.
  • de los Campos et al. (2013). Genome-Enabled Prediction Using the BLR (Bayesian Linear Regression) R-Package. In Genome-Wide Association Studies and Genomic Prediction, edited by Cedric Gondro, Julius van der Werf, and Ben Hayes, 299–320. Methods in Molecular Biology 1019. Humana Press. doi: 10.1007/978-1-62703-447-0_12.
  • Gianola ( 2013). Priors in Whole-Genome Regression: The Bayesian Alphabet Returns. Genetics (May 1): genetics.113.151753. doi:10.1534/genetics.113.151753.
  • Pérez et al. (2010).  Genomic-Enabled Prediction Based on Molecular Markers andPedigree Using the Bayesian Linear Regression Package in R. The Plant Genome Journal 3 (2): 106–116. doi:10.3835/plantgenome2010.04.0005.
  • Vazquez et al. (2012). A Comprehensive Genetic Approach for Improving Prediction of Skin Cancer Risk in Humans. Genetics (October 10). doi:10.1534/genetics.112.141705.
  • de los Campos, et al. (2009). Predicting Quantitative Traits with Regression Models for Dense Molecular Markers and Pedigree. Genetics 182 (1): 375–385. doi: 10.1534/genetics.109.101501

Go to Bayesian Linear Regression (BLR) web page.