Causal inference with simulated genetic data
geneference is an R package for simulating and analysing genetic data for genome wide association studies.
The implementations are primarily based on the work by Hujoel et al (2020), in which the authors propose a method of including family history in a representation of a case-control phenotype. The authors show that their method, named Liability Threshold conditional on Family History, or LT-FH in short, leads to a significant increase in association power in genome wide association studies.
geneference aims to provide a basis for investigating the merits of LT-FH compared to traditional approaches in genome wide association studies. With geneference, the user can control several parameters in the simulations, carry out statistical analysis with different association methods and use plotting functions to visualise certain aspects of the analyses.
geneference can be used as a stand-alone package, but those interested in diving deeper and/or using geneference as a starting point for their own implementations can freely browse our source code at our repository on Github.
geneference was developed as an exam project for the Data Project course, which is part of the Data Science bachelor programme at Aarhus University.
Usage
You can install the development version from GitHub with:
if (!require("devtools")) {
install.packages("devtools")
}
devtools::install_github("FireGutter/geneference")Some of the functions implemented in geneference rely on the software PLINK, which the user will need to install. However, geneference handles the calls to PLINK, meaning that besides installation the user will not need to interact directly with PLINK.
If you are interested in seeing what geneference is capable of, you can read through the introductory vignette, vignette("geneference"). For a complete overview of the functions in geneference, see our package documentation, and for further details on the theoretical background of our implementations, see our vignettes.
References
- Hujoel, M.L.A., Gazal, S., Loh, PR. et al. Liability threshold modeling of case–control status and family history of disease increases association power. Nat Genet 52, 541–547 (2020). https://doi.org/10.1038/s41588-020-0613-6
Additionally, the open source software Inkscape was used together with the hexSticker package to create the logo for geneference.
Finally, we want to thank our supervisor Emil Michael Pedersen for his patience in answering all our questions.
