Skip to contents

Refit CellRank GAM trends for a selected lineage and store the normalized trend matrix and gene modules in `srt@tools$CellRank$trends`.

Usage

RunCellRankTrends(
  srt,
  lineage,
  features = NULL,
  top_n = 500L,
  heatmap_n = 50L,
  time_key = "palantir_pseudotime",
  assay = "RNA",
  layer = "data",
  n_points = 200L,
  norm = TRUE,
  resolution = 0.7,
  max_iter = 1000L,
  spline_order = 3L,
  n_knots = 8L,
  cores = 1L,
  random_state = 0L,
  output_dir = NULL,
  envname = NULL,
  conda = "auto",
  min_expressed_cells = 20L,
  distribution = "gamma",
  link = "log",
  fallback_distribution = "normal",
  verbose = TRUE
)

Arguments

srt

A Seurat object returned by [RunCellRank].

lineage

A lineage name in the stored fate-probability matrix.

features

Optional genes. If `NULL`, positive lineage drivers are used.

top_n

Maximum number of driver genes used for clustering.

heatmap_n

Number of genes retained for the compact heatmap view.

time_key

Metadata column containing pseudotime.

assay

Assay containing the expression layer.

layer

Expression layer used for GAM fitting.

n_points

Number of points used to predict each trend.

norm

Whether to z-normalize each gene trend before clustering.

resolution

Leiden resolution for trend modules.

max_iter

Maximum GAM iterations.

spline_order

Spline order passed to CellRank GAM.

n_knots

Number of GAM knots.

cores

Number of Python workers.

random_state

Reproducibility seed.

output_dir

Optional directory for CSV exports.

envname

Optional Python environment name. NULL uses the current SCOP environment selection.

conda

Conda-compatible executable used by PrepareEnv.

min_expressed_cells

Minimum cells with positive finite expression required for a GAM feature.

distribution

Primary CellRank GAM distribution.

Link function for the primary GAM distribution.

fallback_distribution

Optional distribution used when the primary model has a fatal fit failure; the selected distribution is recorded.

verbose

Whether to print progress messages.

Value

The Seurat object with a `CellRank$trends[[lineage]]` result bundle.