Canonicalize and order an enrichment result data frame — .format_enrichment

Shared post-processing for the ORA / GSEA / CAMERA runners. Each runner hands in a data frame that already has the harmonized columns ChemicalID and pvalue; this helper computes padj, joins chemical metadata, applies a canonical leading-column order, sorts by padj ascending, and drops row names. Centralizing this step keeps the runners small and the public output schema consistent across methods.

Usage

.format_enrichment_result(
  res,
  chemicals_meta,
  pAdjustMethod,
  method,
  rename = NULL,
  drop = NULL
)

Arguments

res: Data frame as returned by an enrichment engine (ora(), gsea(), limma::camera()), still carrying the engine's native column names.
chemicals_meta: Data frame with ChemicalID and ChemicalName columns.
pAdjustMethod: Method passed to p.adjust.
method: Method label (one of "ORA", "GSEA", "CAMERA") stamped into the new Method column so results can be rbind'd across methods without losing provenance.
rename: Named character vector mapping old_engine_colname = "NewCanonicalName". Applied before the metadata merge. Centralizes the engine->canonical-schema mapping so each engine can keep its native column names.
drop: Character vector of engine column names to remove before the merge / sort.

Value

A data frame whose leading columns are ChemicalID, ChemicalName, Method, PValue, PValueAdjusted (whichever of those are present after the rename), followed by the remaining columns in their original order, sorted by PValueAdjusted ascending.