Visualize Output from the LundTaxR Classifier in a Heatmap — plot_classification

Plot heatmap with classification results.

Usage

plot_classification_heatmap(
  these_predictions = NULL,
  this_data = NULL,
  gene_id = "hgnc_symbols",
  subtype_annotation = "5_class",
  this_sample_order = NULL,
  norm = TRUE,
  plot_scores = FALSE,
  plot_signature_scores = FALSE,
  custom_annotation = NULL,
  include_sections = list(early_late_cc = TRUE, luminal_tfs = TRUE, luminal_genes = TRUE,
    fgfr3 = TRUE, circuit = TRUE, tp63 = TRUE, basq = TRUE, keratinization = TRUE, erbb =
    TRUE, adhesion = TRUE, myc = TRUE, neuronal = TRUE),
  show_ann_legend = FALSE,
  show_hm_legend = FALSE,
  ann_height = 0.5,
  plot_title = "My Plot",
  plot_width = 14,
  plot_height = 11,
  plot_font_size = 10,
  plot_font_row_size = 7,
  out_path = NULL,
  out_format = "png",
  col_width = NULL
)

Arguments

these_predictions: Required parameter, should be the output from classify_samples().
this_data: Expression data used for predictions. Required if the output from classify_samples() is run with include_data = FALSE (default).
gene_id: Specify the type of gene identifier used in this_data. Accepted values are; hgnc_symbol (default) or ensembl_gene_id.
subtype_annotation: Can be one of the following; "5_class" (default) or "7_class" annotation.
this_sample_order: Optional, set sample order. Default, samples are split by subtype, and order within each subtype. By default, samples are order by late/early cell cycle ratio (low to high).
norm: Boolean parameter. Set to TRUE (default) to normalize the data into Z-scaled values. If FALSE, data will be row median centered for plotting.
plot_scores: Boolean parameter. Set to TRUE to plot prediction scores for each class. Default is FALSE.
plot_signature_scores: Boolean parameter. Set to TRUE to add signature scores heatmap after the classification heatmap. Default is FALSE.
custom_annotation: Optional HeatmapAnnotation object to add below the existing subtype prediction annotations and score bars. Should be created with get_custom_annotations(). Default is NULL.
include_sections: Named list specifying which heatmap sections to include. Available sections: early_late_cc, luminal_tfs, luminal_genes, fgfr3, circuit, tp63, basq, keratinization, erbb, adhesion, myc, neuronal. Each should be TRUE/FALSE. Default includes all sections (TRUE).
show_ann_legend: Boolean parameter, set to TRUE to show annotation legend (Lund classes). Default is FALSE.
show_hm_legend: Boolean parameter, set to TRUE to show heatmap legend, default is FALSE.
ann_height: Plotting parameter, optional. Annotation height in cm. Default = 8.
plot_title: Plotting parameter. The title for the generated heatmap. Default is "My Plot".
plot_width: This parameter controls the width in inches. Default is 14 (4200 pixels at 300 PPI).
plot_height: This parameter controls the height in pixels. Default is 10 (3000 pixels at 300 PPI)
plot_font_size: Optional parameter to control the size of the font in the generated heatmap. Note, the title of the plot will always be twice that of the set value here (default = 10).
plot_font_row_size: Optional parameter to control the size of the font in the generated heatmap. Note, the title of the plot will always be twice that of the set value here (default = 8).
out_path: Optional, set path to export plot.
out_format: Required parameter if out_path is specified. Can be "png" (default) or "pdf". The user can control the dimensions with plot_width and plot_height.
col_width: Optional parameter to force column width (sample width). Useful for when comparing multiple cohorts and the width of the plot needs to be proportional to the total number of samples. Default is NULL.

Value

Draws heatmap and silently returns the sample order.

Details

This function plots a heatmap including genes and signatures of interest, with prediction results and scores on top. Optionally includes signature scores heatmap.

Examples

#run classifier
sjodahl_classes = classify_samples(this_data = sjodahl_2017, 
                                   log_transform = FALSE, 
                                   adjust = TRUE, 
                                   impute = TRUE, 
                                   include_data = TRUE, 
                                   verbose = FALSE)
                                   
#plot
plot_classification_heatmap(these_predictions = sjodahl_classes, 
                            subtype_annotation = "5_class",
                            plot_scores = FALSE,
                            plot_title = "Classification Results (Sjodahl 2017)", 
                            show_ann_legend = TRUE,
                            ann_height = 0.5)
#> No out_path provided, displaying heatmap in R session...