Skip to content

two-stage analysis of multi-environment trials for genomic selection and GWAS

License

Notifications You must be signed in to change notification settings

jendelman/StageWise

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Jul 24, 2024
ca61ea1 · Jul 24, 2024

History

67 Commits
Jul 24, 2024
Sep 21, 2022
Jul 24, 2024
Feb 16, 2024
Jul 24, 2024
Jul 24, 2024
Sep 28, 2022
Jun 26, 2021
Jul 24, 2024
Jun 22, 2021
Jun 8, 2024
Jul 24, 2024
Mar 22, 2023
Mar 22, 2023

Repository files navigation

R/StageWise

Jeffrey Endelman

Plant breeders typically use multi-environment datasets that are unbalanced with respect to individuals over time and often within year as well. Experimental designs and the best models for micro-environmental variation may also vary with environment, defined as a location x year combination. These complexities can be handled efficiently with a two-stage analysis, in which best linear unbiased estimates (BLUEs) of genotypic value are computed for each environment in Stage 1 and then used as the response variable in Stage 2. To fully utilize the data, however, the variance-covariance matrix of the estimates from Stage 1 should be included in Stage 2 (Piepho et al. 2012; Damesa et al. 2017), which is not possible with most software packages for genomics-assisted breeding.

R package StageWise was developed to provide a simple interface for genomic selection and GWAS based on a fully efficient, two-stage analysis (Endelman 2023). Each phenotypic record is associated with a single genotype id, which is suitable for clonal and inbred lines but not hybrid crops. The ASReml-R package is used for variance component estimation, which requires a license from VSN International.

Software development has been supported by USDA Hatch Project 1013047 and the USDA National Institute of Food and Agriculture (NIFA) Award 2020-51181-32156. The potato datasets were generated with support from NIFA Awards 2016-34141-25707 and 2019-34141-30284, Potatoes USA, the Wisconsin Potato and Vegetable Growers Association, and the University of Wisconsin-Madison.

To install and load the package:

install.packages("devtools")
devtools::install_github("jendelman/StageWise", build_vignettes=FALSE)
library(StageWise)

There are three vignettes in the package:

  • Vignette 1 illustrates analysis of a single trait with a homogeneous GxE model (same genetic correlation between all environments).

  • Vignette 2 illustrates analysis of a single trait with heterogeneous GxE covariance, which is often needed with diverse locations.

  • Vignette 3 illustrates the analysis of correlated traits under a homogeneous GxE model.

For a complete specification of package functions, consult the reference manual.

About

two-stage analysis of multi-environment trials for genomic selection and GWAS

Resources

License

Citation

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages