Skip to contents

Perform cross-validation of merged gene expression data.

Source: R/metapredict_predict.R
metapredictCv.Rd

Run cross-validation to predict a response variable from gene expression data across multiple studies.

Usage

metapredictCv(
  ematMerged,
  sampleMetadata,
  weights,
  alpha,
  nFolds = 10,
  foldid = NA,
  nRepeats = 3,
  yName = "class",
  addlFeatureColnames = NA,
  ...
)

Arguments

ematMerged

matrix of gene expression for genes by samples.

sampleMetadata

data.frame of sample metadata, with rownames corresponding to sample names.

weights

vector of weights.

alpha

vector of values for alpha, the elastic net mixing parameter.

nFolds

number of folds. Ignored, if foldid is not NA.

foldid

vector of values specifying what fold each observation is in.

nRepeats

number of times to perform cross-validation. Ignored, if foldid is not NA.

yName

column in sampleMetadata containing values of the response variable.

addlFeatureColnames

optional vector of column names containing other features to be used for predicting the response variable.

...

Other arguments passed to glmnet::cv.glmnet().

Value

A list of cv.glmnet objects.