PitfallsWPPA corrects the pitfall in GWAS when using WPPA to investigate the genetic association for genomic windows. The work has been published on GSE: https://doi.org/10.1186/s12711-021-00638-x
All the code for the simulation study is shown here. The real dataset is available at http://www.ricediversity.org (Oryza sativa data).
Details of conducting GWAS using WPPA are shown in BayesC.jl. All analyses were performed using the JWAS package, an open-source Julia package for whole-genome analyses based on Bayesian regression methods. The results were analyzed through the original (using a constant threshold value) and modified GWAS (using window specific threshold values), respectively. The modified GWAS function is in GWAS_fun.jl.
A frequentist method of conduction GWAS based on genomic windows, EMMAX, was conducted in R. All the calculations were based on Chen et al. (2017. Genome-wide association analyses based on broadly different specifications for prior distributions, genomic windows, and estimation methods).