Apply the full gate hierarchy to one file — apply_gate

Chain: all events -> leukocytes (scatter) -> single cells -> live -> CD45+. Each step degrades gracefully: a missing viability dye or missing CD45 skips that step with a warning rather than aborting, because a panel that lacks one is still analysable.

Usage

apply_gate_hierarchy(
  rd,
  cofactor,
  cfg = list(),
  control_ref = NULL,
  singlet_k = 3,
  viability_name = NULL,
  cd45_name = "CD45",
  transform = NULL,
  overrides = NULL,
  autofix = FALSE,
  adaptive = FALSE
)

Arguments

rd: The rd.
cofactor: Arcsinh cofactor. See derive_cofactor().
cfg: The cfg. Default list().
control_ref: The control ref.
singlet_k: The singlet k. Default 3.
viability_name: The viability name.
cd45_name: The cd45 name. Default "CD45".
transform: Intensity transform from make_transform(). Defaults to arcsinh with cofactor, which is what every caller got before the transform became selectable.
overrides: Optional per-marker override entries for THIS sample, as returned by indexing the config's sample_overrides block by sample id. See sample_override(). Absent by default, in which case every threshold is derived exactly as before.
autofix: Skip a hierarchy gate whose threshold came from quantile_fallback, carrying its parent through unchanged. Such a gate has its retention decided by the fallback constant rather than by the data, so it keeps a fixed share of the parent whatever the sample holds. FALSE by default, because skipping a gate changes every count in an affected file. See gate_needs_autofix().
adaptive: Sweep the kernel bandwidth and try the tail and Otsu rules before falling back to a quantile, for both hierarchy gates. See best_threshold(). FALSE by default, because it moves thresholds.

Value

list of masks, derived geometry, thresholds, and a tidy counts table