MRAID1sample is an R package for one-sample MRAID. It builds upon the framework that Yuan, et al. (2022) developed. The motivation for this project is to perform one-sample MR in large biobank data sets, such as the UK Biobank.
MRAID(Mendelian Randomization with Automated Instrument Determination),is an R package for efficient statistical inference of two-sample Mendelian Randomization. MRAID takes GWAS summary statistics as inputs to estimate causal effects of one trait on another. MRAID is able to model an initial set of candidate SNP instruments that are in high LD with each other and perform automated instrument selection to identify suitable SNPs to serve as instrumental variables. MRAID simultaneously accounts for both uncorrelated and correlated horizontal pleiotropy, relies on a scalable sampling-based inference algorithm to perform numerical integration, circumventing the difficulty in likelihood function, leading to calibrated p-values that enable reasonably large-scale exposure screening.
This approach is described in,
Yuan Z, Liu L, Guo P, Yan R, Xue F, Zhou X. Likelihood-based Mendelian randomization analysis with automated instrument selection and horizontal pleiotropic modeling. Science Advances 8, eabl5744.
It is easy to install the development version of MRAID package using the 'devtools' package.
# install.packages("remotes")
remotes::install_github("fboehm/MRAID1sample")
The main function in the package is MRAID, you can find the instructions by '?MRAID'.
One simple example to use the package can be found at https://github.com/yuanzhongshang/MRAID/tree/master/example
The MRAID R package is developed by Zhongshang Yuan and Lu Liu.