GWASBrewer: An R Package for Simulating Realistic GWAS Summary Statistics
Author Information
Author(s): Jean Morrison
Primary Institution: Department of Biostatistics University of Michigan
Hypothesis
Can we create a software package that efficiently simulates GWAS summary statistics while maintaining realistic data properties?
Conclusion
GWASBrewer can simulate GWAS summary statistics that closely mimic those generated from individual-level data, but with significantly reduced computational costs.
Supporting Evidence
- GWASBrewer can simulate summary statistics for multiple traits with specified causal relationships.
- Simulations show that GWASBrewer-generated statistics have the same distribution as those from individual-level data.
- The package allows for flexible specification of effect size distributions.
Takeaway
GWASBrewer is a tool that helps scientists create fake data for genetic studies without needing to do a lot of complicated math, making it easier and faster to test their ideas.
Methodology
The package simulates GWAS summary statistics directly from their theoretical distribution, allowing for flexibility in modeling traits and genetic relationships.
Potential Biases
The package does not account for confounding bias from factors such as uncorrected population structure.
Limitations
GWASBrewer currently only supports continuous traits and does not model gene-environment or gene-gene interactions.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website