Run CisTarget motif enrichment analysis — rs

Core Rust function for motif enrichment analysis using recovery curves. Gene sets are processed in parallel; only motifs with an NES at or above nes_threshold are returned.

Usage

rs_cistarget(
  rankings,
  gs_list,
  auc_threshold,
  nes_threshold,
  max_rank,
  method,
  n_mean,
  verbose
)

Arguments

rankings: Integer matrix with motif rankings for genes (genes in rows, motifs in columns). Lower ranks indicate higher regulatory potential.
gs_list: List of integer vectors. Each element contains 1-based indices of genes in the gene set (matching row indices in rankings).
auc_threshold: Integer. Absolute number of top-ranked genes to use for the AUC calculation (e.g., for 5% of 10000 genes, use 500).
nes_threshold: Numeric. Normalised Enrichment Score threshold for filtering significant motifs.
max_rank: Integer. Maximum rank to consider for the recovery curves (at most nrow(rankings)).
method: String. Recovery curve calculation method, one of c("approx", "icistarget"). Anything else falls back to "approx".
n_mean: Integer. Window size for the smoothing in the approximate method.
verbose: Boolean. Report progress per decile of gene sets.

Value

List of lists, one per gene set, each containing

motif_idx - 1-based column index of the motif in rankings.
nes - Normalised enrichment score.
auc - Area under the recovery curve.
rank_at_max - Rank at which the leading edge is reached.
n_enriched - Number of genes in the leading edge.
leading_edge - List of 1-based row indices of the leading edge genes, one element per motif.