GSbench: Benchmarking Genomic Selection and Machine-Learning Prediction Models

A unified interface to fit, cross-validate and benchmark genomic prediction models from SNP marker data. It implements genomic best linear unbiased prediction (GBLUP) and ridge-regression BLUP in base R, and offers a common interface to machine-learning predictors (elastic net, random forest and gradient boosting) through optional packages, together with a stacked ensemble. Cross-validation uses breeding-relevant schemes and reports prediction accuracy honestly, so models can be compared fairly. The genomic relationship matrix follows VanRaden (2008) <doi:10.3168/jds.2007-0980>; the mixed-model solver follows Endelman (2011) <doi:10.3835/plantgenome2011.08.0024>; the genomic-selection framework follows Meuwissen, Hayes and Goddard (2001) <doi:10.1093/genetics/157.4.1819>.

Version:	0.1.0
Depends:	R (≥ 4.1.0)
Imports:	graphics, stats, withr
Suggests:	rrBLUP, glmnet, ranger, xgboost, testthat (≥ 3.0.0), knitr, rmarkdown, spelling
Published:	2026-06-30
DOI:	10.32614/CRAN.package.GSbench (may not be active yet)
Author:	Muhammad Farooqi [aut, cre]
Maintainer:	Muhammad Farooqi <mqfarooqi at gmail.com>
BugReports:	https://github.com/mqfarooqi1/GSbench/issues
License:	MIT + file LICENSE
URL:	https://github.com/mqfarooqi1/GSbench
NeedsCompilation:	no
Language:	en-GB
Materials:	README, NEWS
CRAN checks:	GSbench results

Documentation:

Reference manual:	GSbench.html , GSbench.pdf
Vignettes:	Benchmarking genomic prediction models with GSbench (source, R code)

Downloads:

Package source:	GSbench_0.1.0.tar.gz
Windows binaries:	r-devel: not available, r-release: not available, r-oldrel: not available
macOS binaries:	r-release (arm64): GSbench_0.1.0.tgz, r-oldrel (arm64): not available, r-release (x86_64): GSbench_0.1.0.tgz, r-oldrel (x86_64): GSbench_0.1.0.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=GSbench to link to this page.