The model is saved in a defined directory.
build_model_scaden(
single_cell_object,
cell_type_annotations,
bulk_gene_expression,
model_path = NULL,
temp_dir = NULL,
batch_size = 128,
learning_rate = 1e-04,
steps = 5000,
var_cutoff = NULL,
cells = 100,
samples = 1000,
dataset_name = "scaden",
verbose = FALSE
)A matrix with the single-cell data. Rows are genes, columns are samples. Row and column names need to be set.
A vector of the cell type annotations. Has to be in the same order as the samples in single_cell_object.
A matrix of bulk data. Rows are genes, columns are samples. Row and column names need to be set.
Path where model directory should be created (optional).
The temporary directory to use for the computations (optional)
Training of model: Batch size to use for training (default: 128).
Training of model: Learning rate used for training (default: 1E-4).
Training of model: Number of training steps (default: 5000).
Training data processing: Filter out genes with a variance less than the specified cutoff. A low cutoff is recommended,this should only remove genes that are obviously uninformative. (default NULL).
Bulk simulation: Number of cells per sample (default: 100).
Bulk simulation: Number of samples to simulate (default: 1000).
Bulk simulation: Name of simulated dataset (default scaden).
Whether to produce an output on the console (default: false).
The path to the scaden model.