Welcome

Welcome to GWASBrewer, a flexible tool for simulating realistic GWAS summary statistics for one or many traits from a wide range of scenarios.

This package was previously called simGWAS. Congratulations on finding our new home.

Simulate Realistic GWAS Summary Statistics for One or Multiple Traits

The GWASBrewer package simulates GWAS summary statistics. The main function in the package is sim_mv. Get started with the "Simulating Data" vignette.

Briefly, GWASBrewer can simulate data with the following features.

• Data an be produced for continuous traits with user supplied linear causal relationships.
• GWAS for multiple traits may have overlapping samples.
• Data can be generated with or without LD. One realistic LD pattern is supplied as built-in data.
• GWAS for the same trait can be replicated with different sample sizes, LD patterns, and allele frequencies (see the "Resampling and Re-Scaling.." vignette)

Data generated by GWASBrewer can be used for testing a variety of methods including heritability estimation, Mendelian randomization, genetic correlation estimation, colocalization, fine mapping etc.

This package is under active development. Some features that may be added in the future include

• Support for binary traits
• Ability to add confounding effects

Installation

devtools::install_github("jean997/GWASBrewer",  build_vignettes = TRUE)
browseVignettes("GWASBrewer")

or install without vignettes,

devtools::install_github("jean997/GWASBrewer",  build_vignettes = FALSE)

The vignettes require two packages that will not be installed automatically. The "Simulating Data" vignette requires DiagrammeR (use install.packages).

Generating individual level data using resample_inddata (see "Resampling and Rescaling..." vignette) requires the hapsim package. This package was removed from CRAN in June 2025 but can be installed using devtools::install_github("CRAN/hapsim").