Canonical signature scoring interface

Canonical signature scoring interface

Usage

score_signature(
  object = NULL,
  expr = NULL,
  meta = NULL,
  geneset,
  geneset_source = c("auto", "builtin", "list", "gmt", "data.frame", "msigdb", "go",
    "kegg", "reactome", "wikipathways", "biocarta"),
  species = "Homo sapiens",
  collection = "H",
  subcollection = NULL,
  ontology = c("BP", "MF", "CC"),
  seurat = TRUE,
  assay = NULL,
  layer = NULL,
  slot = NULL,
  method = c("rank", "mean", "zscore", "scaled_mean", "robust_mean", "ensemble",
    "AddModuleScore", "UCell", "AUCell", "ssGSEA", "GSVA", "singscore"),
  min_genes = 5,
  max_genes = 500,
  auc_max_rank = 0.05,
  ensemble_methods = c("rank", "zscore", "mean"),
  ensemble_combine = c("mean", "median"),
  ensemble_standardize = c("zscore", "rank"),
  ensemble_weights = NULL,
  method_params = list(),
  verbose = TRUE
)

Arguments

object

Seurat object when seurat = TRUE.

expr

Expression matrix or dgCMatrix when seurat = FALSE.

meta

Metadata data.frame for matrix mode.

geneset

Geneset input. Also supports signed signatures: list(signatureA = list(up = c(...), down = c(...))).

geneset_source

Geneset source (auto, builtin, list, gmt, data.frame, msigdb, go, kegg, reactome, wikipathways, biocarta).

species

Species label for source-aware geneset loading.

collection

Collection parameter for source-aware geneset loading.

subcollection

Subcollection parameter for source-aware geneset loading.

ontology

GO ontology (BP, MF, CC) for GO source.

seurat

Logical input mode.

assay

Seurat assay name.

layer

Seurat v5 layer name. If NULL, tries data then counts.

slot

Legacy Seurat slot fallback.

method

Scoring method. Canonical methods include: rank, mean, zscore, scaled_mean, robust_mean, ensemble, AddModuleScore, UCell, AUCell, ssGSEA, GSVA, singscore.

min_genes

Minimum genes per pathway.

max_genes

Maximum genes per pathway.

auc_max_rank

AUC top-rank proportion used by AUCell.

ensemble_methods

Methods used by ensemble.

ensemble_combine

Ensemble combine strategy.

ensemble_standardize

Harmonization before ensemble aggregation: zscore (recommended) or rank.

ensemble_weights

Optional named numeric vector of per-method weights.

method_params

Optional named list for method-specific parameters. Supported keys include: auc_max_rank (AUCell), alpha or ssgsea_alpha (ssGSEA), nbin/ctrl/seed (AddModuleScore), kcdf (GSVA), ucell_max_rank (UCell).

verbose

Whether to print messages.

Value

An object of class gleam_score.

Method Guide

Method	Best for	Notes
rank	Fast robust baseline	Rank aggregation concept aligned with singscore
UCell	Large single-cell data	Mann-Whitney U-based robust scoring
AUCell	Rank-enrichment scoring	SCENIC/AUCell implementation
AddModuleScore	Seurat-native workflows	Signature minus control-bin expression
ssGSEA	Bulk-like enrichment workflows	GSVA package ssGSEA implementation
GSVA	Bulk expression matrices	GSVA kernel-based pathway variation
mean	Simple signatures	Mean expression aggregation
zscore	Cross-signature comparability	Per-gene z-score then aggregate
robust_mean	Outlier-prone data	Median/MAD-scaled robust aggregation
ensemble	Consensus scoring	Harmonize each method (zscore or rank) before averaging

Method Parameters (method_params)

Supported keys include: auc_max_rank (AUCell), alpha or ssgsea_alpha (ssGSEA), nbin/ctrl/seed (AddModuleScore), kcdf (GSVA), ucell_max_rank (UCell). For ensemble, use ensemble_methods, ensemble_standardize, and optional ensemble_weights.