Skip to contents

Run CellTypist cell type annotation

Source: R/RunCellTypist.R
RunCellTypist.Rd

CellTypist is an automated cell type annotation tool for scRNA-seq datasets based on logistic regression classifiers. This function runs CellTypist annotation on a Seurat object or AnnData object.

Usage

RunCellTypist(
  srt = NULL,
  adata = NULL,
  assay = "RNA",
  layer = "data",
  model = "Immune_All_Low.pkl",
  mode = "best match",
  p_thres = 0.5,
  majority_voting = FALSE,
  over_clustering = NULL,
  min_prop = 0,
  use_GPU = FALSE,
  insert_labels = TRUE,
  insert_conf = TRUE,
  insert_conf_by = "predicted_labels",
  insert_prob = FALSE,
  insert_decision = FALSE,
  prefix = "celltypist_",
  return_seurat = !is.null(srt),
  verbose = TRUE
)

Arguments

srt

A Seurat object. If provided, adata will be ignored.

adata

Optional AnnData object used as input.

assay

Which assay to use.

layer

Assay layer to use.

model

Model name or path. Supports three formats:

  1. Model name (e.g., "Immune_All_Low.pkl"): automatically searched in ~/.celltypist/data/models/

  2. Full path (contains /): use the provided path directly

  3. NULL: use default model

  4. A summary list returned by TrainCellTypist(): use its model_path

mode

Prediction mode: "best match" or "prob match".

p_thres

Probability threshold for "prob match" mode.

majority_voting

Whether to use majority voting.

over_clustering

Over-clustering result. Can be:

  • String: column name in Seurat metadata or AnnData obs

  • Vector: over-clustering labels

  • NULL: use heuristic over-clustering

min_prop

Minimum proportion for majority voting.

use_GPU

Whether to use GPU for over-clustering.

insert_labels

Whether to insert predicted labels.

insert_conf

Whether to insert confidence scores.

insert_conf_by

Which prediction type to base confidence on.

insert_prob

Whether to insert probability matrix.

insert_decision

Whether to insert decision matrix.

prefix

Prefix for inserted columns.

return_seurat

Whether to return a Seurat object instead of an anndata object.

verbose

Whether to print the message. Default is TRUE.

Value

An AnnData object or a Seurat object depending on the return_seurat argument.

See also

CellTypistModels, TrainCellTypist, RunSingleR, RunScmap

Examples

if (FALSE) { # \dontrun{
data(pancreas_sub)
pancreas_sub <- RunStandardWorkflow(pancreas_sub)
pancreas_sub <- RunCellTypist(
  pancreas_sub,
  model = "Developing_Mouse_Brain.pkl"
)
CellDimPlot(
  pancreas_sub,
  group.by = "celltypist_predicted_labels",
  legend.position = "none"
)

# Use prob match mode
pancreas_sub <- RunCellTypist(
  pancreas_sub,
  model = "Developing_Mouse_Brain.pkl",
  mode = "prob match",
  p_thres = 0.5
)

# Use majority voting
pancreas_sub <- RunCellTypist(
  pancreas_sub,
  model = "Developing_Mouse_Brain.pkl",
  majority_voting = TRUE
)
} # }