cell_annotation_diagnostic_plots_fn: Generate diagnostic plots for cell annotation.

In bioinfocz/scdrake: A pipeline for droplet-based single-cell RNA-seq data secondary analysis implemented in the drake Make-like toolkit for R language

Generate diagnostic plots for cell annotation.

Description

Three list and character vector columns are added to cell_annotation:

• score_heatmaps: heatmaps of per-cell label scores created for each clustering (as column annotation) (details). Generated by SingleR::plotScoreHeatmap()

• marker_heatmaps: NULL if cell_annotation$train_params$de is not "de", otherwise heatmap for each label containing top upregulated markers from pairwise t-tests (details). Number of top markers is specified in cell_annotation$diagnostics_params$heatmap_n_top_markers

• delta_distribution_plot: violin plots (in one figure / object) of per-cell deltas for each label. Deltas are differences between the score for the assigned label and the median across all labels for each cell (details)

Columns with output PDF files are named as score_heatmaps_out_file etc.

Usage

cell_annotation_diagnostic_plots_fn(
  cell_annotation,
  cell_data,
  sce,
  base_out_dir,
  do_heatmaps = FALSE,
  cluster_cols_regex = "^cluster_"
)

Arguments

cell_annotation	(input target) A tibble.
cell_data	(input target) A DataFrame.
sce	A SingleCellExperiment object.
base_out_dir	A character scalar: path to output directory under which will be for each reference dataset its diagnostic plots saved in.
do_heatmaps	A logical scalar: if TRUE, generate score heatmaps.
cluster_cols_regex	A character scalar: regex to match columns in colData() with cluster labels. Those columns will be used in annotation score heatmaps.

Value

A tibble. Output target: cell_annotation_diagnostic_plots

bioinfocz/scdrake documentation built on Sept. 19, 2024, 4:43 p.m.