This package provides functions to detect and correct for batch effects in DNA methylation data. The core function for the data imputation is based on latent factor models and can also be used to predict missing values in any other matrix containing real numbers.
# Installation from Bioconductor
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("BEclear")# Installation of the development version from GitHub
if (!requireNamespace("devtools", quietly = TRUE))
install.packages("devtools")
devtools::install_github("uds-helms/BEclear", build_opts = c())For example code you can either read the vignette on Bioconductor. Or you can open it local, after you installed the BEclear package with following command:
browseVignettes("BEclear")- Treating Data-Sets Without Batches
- Make Saving Temporary Data Optional
- Bias Modelling
- ALS
- Test For Convergence Durin Epochs
- Testing BEclear on other data-sets
- After Merging Blocks Continue GD
If you have a suggestion you could either open an issue or write an email under the following address: Livia.Rasp at gmail.com
Akulenko, R., Merl, M., & Helms, V. (2016). BEclear: Batch effect detection and adjustment in DNA methylation data. PLoS ONE, 11(8), 1–17. https://doi.org/10.1371/journal.pone.0159921