predict.beefedUpSNC: fSVA and Nearest Shrunken Centroid prediction

In MalteThodberg/BEEF: Batch Effect Estimation with fSVA (Frozen Surrogate Variable Analysis)

Description

Predict labels or posterior probabilities for new samples.

Usage

## S3 method for class 'beefedUpSNC'
predict(object, x, threshold = NULL, type = "class",
  ...)

Arguments

object	beefedUpSNC object.
x	matrix: Normalized and log-transformed expression values (samples in rows, features in columns).
threshold	numeric: Amount of centroid shrinkage (controls how many features are retained in the model).
type	character: Either "class" for labels or "posterior" for class probabilities.
...	additional arguments, currently not used.

Value

factor if type = "class" or matrix of class probabilities if type = "posterior"

Examples

# ADD_EXAMPLES_HERE

MalteThodberg/BEEF documentation built on May 7, 2019, 2:09 p.m.