Process Training Data for TorchDecon

Preprocess simulated bulk data for model training. This includes log transformation, scaling, and gene filtering based on variance and intersection with prediction data.

Usage

ProcessTrainingData(
  simulation,
  prediction_data = NULL,
  var_cutoff = 0.1,
  scaling = c("log_min_max", "log_zscore", "none"),
  verbose = TRUE
)

Arguments

simulation

A TorchDeconSimulation object from SimulateBulk, or a list/matrix of bulk counts.

prediction_data

Matrix or data frame of bulk expression data for prediction (genes in rows, samples in columns). Used to find common genes.

var_cutoff

Numeric. Filter out genes with variance below this threshold. Default is 0.1.

scaling

Character. Scaling method to use. One of "log_min_max" (default), "log_zscore", or "none".

verbose

Logical. Print progress messages. Default is TRUE.

Value

A list containing:

X

Processed expression matrix (samples x genes), ready for training

Y

Cell type fractions matrix (samples x cell types)

genes

Character vector of genes used (signature genes)

celltypes

Character vector of cell type names

scaling

Scaling method used

scaling_params

Parameters for scaling (for applying to new data)

Details

The preprocessing pipeline:

1. Find common genes between training and prediction data

2. Filter genes by variance threshold

3. Apply log2(x + 1) transformation

4. Apply sample-wise min-max scaling (or z-score)

Examples

if (FALSE) { # \dontrun{
# Basic processing
processed <- ProcessTrainingData(simulation, prediction_data = bulk_data)

# Custom variance cutoff
processed <- ProcessTrainingData(
  simulation,
  prediction_data = bulk_data,
  var_cutoff = 0.05,
  scaling = "log_min_max"
)
} # }