oplsda {ASICS} | R Documentation |
SummarizedExperiment
objectPerform an OPLS-DA with the function of the ropls
package on a
SummarizedExperiment
object obtained with the
formatForAnalysis
function
oplsda( analysis_data, condition, cross.val = 1, thres.VIP = 1, type.data = "quantifications", seed = 12345, ... )
analysis_data |
A |
condition |
The name of the design variable (two level factor) specifying the response to be explained. |
cross.val |
Number of cross validation folds. |
thres.VIP |
A number specifying the VIP threshold used to identify influential variables. |
type.data |
Type of data used for the analyses (e.g.,
|
seed |
Random seed to control randomness of cross validation folds. |
... |
Further arguments to be passed to the function
|
A S4 object of class AnalysisResults containing OPLS-DA results.
Trygg, J. and Wold, S. (2002). Orthogonal projections to latent structures (O-PLS). Journal of Chemometrics, 16(3), 119–128.
Thevenot, E.A., Roux, A., Xu, Y., Ezan, E., Junot, C. 2015. Analysis of the human adult urinary metabolome variations with age, body mass index and gender by implementing a comprehensive workflow for univariate and OPLS statistical analyses. Journal of Proteome Research. 14:3322-3335.
# Import quantification results if (require("ASICSdata", quietly = TRUE)) { quantif_path <- system.file("extdata", "results_ASICS.txt", package = "ASICSdata") quantification <- read.table(quantif_path, header = TRUE, row.names = 1) # Import design design <- read.table(system.file("extdata", "design_diabete_example.txt", package = "ASICSdata"), header = TRUE) design$condition <- factor(design$condition) # Create object for analysis and remove features with more than 25% of # zeros analysis_obj <- formatForAnalysis(quantification, zero.threshold = 25, design = design) res_oplsda <- oplsda(analysis_obj, "condition", orthoI = 1) }