Predict Tumor Grade — int_predict

Apply SwitchBox Classifier to Gene Expression Data

Usage

int_predict_grade(
  data,
  classifier,
  grade_threshold = 0.5,
  grade_labels = NULL,
  verbose = TRUE
)

Arguments

data: Gene expression matrix (genes × samples). Row names must be gene identifiers.
classifier: SwitchBox classifier object with $TSPs, $score components
grade_threshold: Numeric cutoff for binary classification (default 0.5)
grade_labels: Character vector of length 2: c("low_label", "high_label")
verbose: Print diagnostic information

Value

Data frame with sample IDs as row names and two columns:

prediction_score: Numeric scores (0-1 scale)
predicted_class: Character classification based on threshold

Details

Internal function called by int_calc_signatures(). Not meant for out of package use. This function applies a pre-trained SwitchBox classifier to gene expression data to predict molecular tumor grades. The classifier uses Top Scoring Pairs (TSPs) methodology, comparing expression levels between gene pairs to generate predictions. Each rule evaluates whether gene1 > gene2, and satisfied rules contribute their weights to the final score.

Examples

if (FALSE) { # \dontrun{
# No examples provided
} # }