Generate a heatmap based on transcriptomic data for a basal-like/classical gene list

The function selects the expression values related to the gene list passed in argument. Then, the function adds one to the expression values before log transforming the values. Then, the function normalizes the data so that each row has mean 0 and variance 1. Finally, the function generates a heatmap using input parameters.

createHeatmap(
  gene_list,
  rna_data,
  gene_column = "GENE",
  name = "Expression",
  clustering_distance_rows = c("euclidean", "maximum", "manhattan", "canberra",
    "binary", "minkowski", "pearson", "spearman", "kendall"),
  clustering_distance_columns = c("euclidean", "maximum", "manhattan", "canberra",
    "binary", "minkowski", "pearson", "spearman", "kendall"),
  show_column_dend = TRUE,
  show_row_dend = TRUE,
  cluster_columns = TRUE,
  cluster_rows = TRUE,
  log_data = TRUE,
  heatmap_legend_param = gpar(title_gp = gpar(col = "black", fontsize = 11, fontface =
    "bold")),
  ...
)

Arguments

gene_list

a data.frame containing the gene list used for the basal-like/classical classification. The data.frame must contain two columns: SYMBOL and Class. The SYMBOL contains the gene symbol while the Class column contains its classification. The classification can be Classical or Basal-like.

rna_data

data.frame containing the expression level for the samples. Each column must represent a sample. An extra column for the gene name must be present. The name of this column must correspond to the parameter gene_column.

gene_column

a character string representing the name of the column in the rna_data parameter that contains the gene names. Default: GENE.

name

a character string representing the title of the heatmap legend. Default: "Expression".

clustering_distance_rows

a character string representing the method uses to cluster the rows of the heatmap. The available choices are: "euclidean", "maximum", "manhattan", "canberra", "binary", "minkowski", "pearson", "spearman" and "kendall". Default: "euclidean".

clustering_distance_columns

a character string representing the method uses to cluster the columns of the heatmap. The available choices are: "euclidean", "maximum", "manhattan", "canberra", "binary", "minkowski", "pearson", "spearman" and "kendall". Default: "euclidean".

show_column_dend

a codelogical indicating if the column dendrogram should be shown. Default: TRUE.

show_row_dend

a codelogical indicating if the row dendrogram should be shown. Default: TRUE.

cluster_columns

a codelogical indicating if the columns should be clustered. Default: TRUE.

cluster_rows

a codelogical indicating if the rows should be clustered. Default: TRUE.

log_data

a logical indicating if the data should be log transformed. If so, the data will be log2(value + 1) transformed. Default: TRUE.

heatmap_legend_param

an object of class 'gpar'. Default: gpar(title_gp=gpar(col="black", fontsize=11, fontface="bold")).

...

further arguments passed to ComplexHeatmap::Heatmap() function.

Value

an object of class ComplexHeatmap that represent a heatmap based on transcriptomic data for a specific basal-like/classical gene list.

See also

The default method ComplexHeatmap::Heatmap().

Author

Astrid Deschênes

Examples

## Load Moffitt et al 2015 gene list
data("Moffitt_2015")

## Load demo PDAC TCGA expression data
data("demo_PDAC_TCGA")

## Create heatmap with default parameters
## The column in the rna_data data.frame is the default name "GENE"
createHeatmap(gene_list=Moffitt_2015, rna_data=demo_PDAC_TCGA)