Package index • cytoGateR

Marker-aware gating

Build interpretable cell-type scores, gates, and gating trees.

fit_gmm_2(): Fit a 2-component Gaussian mixture model
fit_marker_stats(): Fit per-marker 2-GMM statistics for soft gating
gmm_equal_posterior_cutoff(): Compute the cutoff where posteriors are equal
marker_separability(): Marker separability
neg_penalty(): Negative marker penalty
score_marker_logistic(): Logistic marker score
score_marker_rank(): Rank based marker score
calculate_soft_scores(): Calculate soft gating probabilities per cell type
assign_soft_labels(): Assign hard labels from probability matrix
run_soft_gating(): Run soft-gating pipeline end-to-end
run_tree_gating(): Run tree-based soft gating on a SpatialExperiment
build_fullcoverage_tree(): Build full-coverage gating tree
collect_path_scores(): Collect scores along a tree path
tree_prob(): Tree probability for one cell

Select confident references and classify cells that remain unresolved.

apply_cutoff_labels(): Apply cutoff-based labels to a result list
assign_confident_labels(): Assign labels with uncertainty handling
compute_custom_labels(): Flag confident cells from a probability matrix
custom_labels(): Add custom labels to a SpatialExperiment
prob_mad_cutoff(): Median + MAD cutoff helper
prob_quantile_cutoff(): Default quantile-based cutoff
train_custom_knn(): Train kNN Reference for Cell Type Prediction
predict_unknown_with_knn(): Predict and fill unknown labels using a weighted kNN model
predict_wknn_multi(): Multi-Metric Weighted kNN
train_custom_randomforest(): Train a custom random forest with consensus label cleaning
predict_unknown_with_randomforest(): Predict and fill unknown labels using a trained random forest model
train_custom_dl(): Train a custom deep-learning classifier with consensus label cleaning
predict_unknown_with_dl(): Predict unknown cells using a trained deep-learning model
build_lineage_hierarchy(): Build Lineage Hierarchy from Core Cells
build_hierarchical_reference(): Build Hierarchical Reference with Flexible Marker Selection
predict_hierarchical_knn_recursive(): Recursive Hierarchical kNN with BiocParallel Ensemble

Quantify agreement, classification performance, and spatial uncertainty.

calculate_f1(): Calculate F1 Score per Cell Type
calculate_uncertainty(): Calculate Full Uncertainty Suite with Sample Grouping
calculate_spatial_prior_labels(): Apply Protected Spatial Priors to kNN Probabilities
class_metrics_from_fit(): Derive per-class metrics from a fitted model
label_agreement_rates(): Agreement rates between two label columns
label_confusion_matrix(): Confusion matrix between two label columns
probability_label_matrix(): Build a binary label matrix from probabilities
rf_metric_table(): Tabular and text summaries of RF performance
rf_metric_text(): Text summary of random forest performance

Inspect marker distributions, spatial labels, probabilities, and trees.

plot_celltype_tree(): Plot a cell-type gating tree
plot_class_metrics(): Plot per-class precision, recall, and F1
plot_confusion_matrix(): Plot a confusion matrix heatmap
plot_ct_marker_intensity(): Plot marker intensity with optional GMM fit
plot_label_agreement(): Plot agreement between two label columns
plot_label_cardinality(): Barplot of positive label counts per cell
plot_label_confusion_matrix(): Plot confusion matrix between two label columns
plot_label_counts(): Plot counts of a label column
plot_label_dotplot(): Plot label comparison dotplot
plot_labelled_cells(): Plots the spatial arrangement of classified cells
plot_marker_density(): Plot marker density across images
plot_marker_priority_tree(): Plot a marker priority "tree" for a given cell type (weights order)
plot_probability_map(): Plot spatial probability map for a cell type
plot_pseudobulk_heatmap(): Plot pseudobulk marker heatmap
plot_rand_cell_probs(): Plot the cell type probabilities of a random sample of cells
plot_score_hist(): Plot cell-type probability histograms
print_celltype_tree(): Print a cell-type gating tree
tree_to_df(): Convert a tree to a tibble representation