This repository is based on the following paper (currently in preparation):
shinyGEM: A computationally efficient epigenome wide statistical analysis software, with a focus on batch effects by Gabrielle O’Connor, Phillip E Melton, Rae-Chi Huang, Ashleigh Lin, Gopalan Nair, and Sebastian Rauschert.
In this repository we develop a R package for fast linear modelling of genetic/epigenetic or any other large omic data. This project is based on the
GEM (gene, environment, methylation) package, developed by Hong P. et al.(2016).
The GEM package uses functionalities from the MatrixEQTL package, developed by Shabalin A (2012).
A publication, outlining further testing, development and reasons for the development of this package is in preparation.
The application and a practical example for the usage this package can be found on the instruction page.
We aimed to build an easy to use package for wider application in the research community, enabling non-bioinformaticians to perform epigenome/genome wide association studies.
The main functions in this package are:
shinyGEM_Emodel()shinyGEM()VolcanoPlot()
So far, the package was tested with Illumina HumanMethylation450 data (~450k variables, 8.51GB). For this kind of data, we recommend at least:
- A Windows/Mac/Linux machine with at least 16GB of RAM
- RStudio > 3.5
