Applies the confidence-based labelling scheme to a probability matrix and stores the resulting labels in `colData(spe)`.
Usage
custom_labels(
spe,
prob_mat,
colname = "custom_label",
base_thresh = 0.4,
quantile_cut = 0.75,
iqr_mult = 1.5,
unknown_label = "Unknown",
flag_fn = NULL
)Arguments
- spe
A `SpatialExperiment` or `SingleCellExperiment`.
- prob_mat
Numeric probability matrix (cells x lineages). Rows must correspond to `colnames(spe)`; if rownames are present they are checked.
- colname
Name of the output column in `colData` (default "custom_label").
- base_thresh
Minimum allowable threshold (default 0.4).
- quantile_cut
Quantile used to set the adaptive threshold (default 0.75).
- iqr_mult
Multiplier applied to the IQR (default 1.5).
- unknown_label
Label assigned to uncertain cells (default "Unknown").
- flag_fn
Optional function taking a probability vector and returning a logical indicating confidence; defaults to the internal Tukey-style rule.
Examples
data("cytoGateR_example", package = "cytoGateR")
image_ids <- SummarizedExperiment::colData(cytoGateR_example)$image_name
spe <- cytoGateR_example[, image_ids == image_ids[1L]]
prob_cols <- grep(
"^KNN_P_",
names(SummarizedExperiment::colData(spe)),
value = TRUE
)
prob_mat <- as.matrix(SummarizedExperiment::colData(spe)[, prob_cols])
rownames(prob_mat) <- colnames(spe)
colnames(prob_mat) <- sub("^KNN_P_", "", colnames(prob_mat))
spe <- custom_labels(
spe,
prob_mat,
flag_fn = function(x) max(x) >= 0.8
)
table(SummarizedExperiment::colData(spe)$custom_label)
#>
#> Bcell BnTcell CD4 CD8 Myeloid Neutrophil
#> 401 976 505 381 210 30
#> Plasma_cell Stroma Treg Tumor Unknown
#> 17 347 110 70 1316