batchCorrection {MultiBaC} | R Documentation |
This function performs the ARSyNbac correction [1] for each omic contained in mulBatchDesign input object.
batchCorrection(mbac, multiBatchDesign, Interaction = FALSE, Variability = 0.9)
mbac |
mbac object generated by createMbac. PLS models slot must be present. |
multiBatchDesign |
A list containing the original and predicted omic for each batch. All omics must be present in every batch. Output object of genMissingOmics function |
Interaction |
Logical. Whether to model the interaction between experimental factors and bacth factor in ARSyN models. By default, FALSE. |
Variability |
From 0 to 1. Minimum percent of data variability that must be explained for each ARSyN model. By default, 0.90. |
Custom mbac object. Elements in a mbac object:
ListOfBatches: A list of MultiAssayExperiment objects (one per batch).
commonOmic: Name of the common omic between the batches.
CorrectedData: Same structure than ListOfBatches but with the corrected data instead of the original.
PLSmodels: PLS models created during MultiBaC method performance (one model per non-common omic data type).
ARSyNmodels: ARSyN models created during MultiBaC performance (one per omic data type).
InnerRelation: Table of class data.frame containing the inner correlation (i.e. correlation between the scores of X (t) and Y (u) matrices) for each PLS model across all components.
[1] Nueda MJ, Ferrer A, Conesa A. ARSyN: A method for the identification and removal of systematic noise in multifactorial time course microarray experiments. Biostatistics. 2012;13(3):553-566. doi:10.1093/biostatistics/kxr042
data('multiyeast') my_mbac <- createMbac (inputOmics = list(A.rna, A.gro, B.rna, B.ribo, C.rna, C.par), batchFactor = c("A", "A", "B", "B", "C", "C"), experimentalDesign = list("A" = c("Glu+", "Glu+", "Glu+", "Glu-", "Glu-", "Glu-"), "B" = c("Glu+", "Glu+", "Glu-", "Glu-"), "C" = c("Glu+", "Glu+", "Glu-", "Glu-")), omicNames = c("RNA", "GRO", "RNA", "RIBO", "RNA", "PAR"), commonOmic = "RNA") my_mbac_2 <- genModelList (my_mbac, test.comp = NULL, scale = FALSE, center = TRUE, crossval = NULL, showinfo = TRUE) multiBatchDesign <- genMissingOmics(my_mbac_2) my_finalwise_mbac <- batchCorrection(my_mbac_2, multiBatchDesign = multiBatchDesign, Interaction = FALSE, Variability = 0.9)