Usage
## S3 method for class 'data.frame':
biosign(x, ...)
## S3 method for class 'matrix':
biosign(x,
y,
methodVc = c("all", "plsda", "randomforest", "svm")[1],
bootI = 50,
pvalN = 0.05,
permI = 1,
fixRankL = FALSE,
printL = TRUE,
plotL = TRUE,
.sinkC = NULL,
...)
Arguments
x
Numerical data frame or matrix (observations x variables); NAs are allowed for
PLS-DA but for SVM, samples with NA will be removed
y
Factor: Two-level factor corresponding to the class labels
methodVc
Character vector: Either one or all of the following classifiers:
Partial Least Squares Discriminant Analysis ('plsda'), or Random Forest ('randomforest'), or Support Vector Machine ('svm')
bootI
Integer: Number of bootstaps for resampling
pvalN
Numeric: To speed up the selection, only variables which
significantly improve the model up to two times this threshold (to
take into account potential fluctuations) are computed
permI
Integer: Random permutation are used to assess the significance of each new variable included into the model (forward selection)
fixRankL
Logical: Should the initial ranking be computed with the full model only,
or as the median of the ranks from the models built on the sampled dataset?
printL
Logical: Should informations regarding the data set and the model be
printed? [default = TRUE]
plotL
Logical: Should the 'summary' plot be displayed? [default = TRUE]
.sinkC
Character: Name of the file for R output diversion [default = NULL:
no diversion]; Diversion of messages is required for the integration
into Galaxy