Evaluate candidate WGCNA soft-thresholding powers — coexpr_pick

Runs WGCNA's scale-free topology diagnostics directly from an MAE assay and applies the same automatic sample/feature quality filtering used by coexpr_wgcna(). The returned fit indices support, but do not replace, the scientific choice of a soft-thresholding power.

Usage

coexpr_pick_power(
  x,
  experiment,
  assay,
  powers = c(seq_len(10L), seq(12L, 20L, by = 2L)),
  top_n = NULL,
  network_type = "signed",
  r_squared = 0.85,
  verbose = 0,
  ...
)

Arguments

x: A MultiAssayExperiment object.
experiment: Optional experiment name.
assay: Assay name or one-based assay index.
powers: Positive candidate powers.
top_n: Optional number of most variable genes.
network_type: WGCNA network type. Use the same value for module preservation.
r_squared: Target scale-free topology fit passed as RsquaredCut.
verbose: Backend verbosity.
...: Additional arguments passed to WGCNA::pickSoftThreshold().

Value

The native pickSoftThreshold() list, augmented with bulkMAEQuality, bulkMAERemovedSamples, and bulkMAERemovedFeatures elements.