Create Gene Exploration Plot

Plots, for every gene of the experiment that is a gene node of the network, the experimental p-value against the tKOI false discovery rate (FDR), with up- and down-regulated genes in separate panels. Genes supported by both the experiment and the network appear in the top right of each panel.

Usage

make_gene_exploration_plot(
  tkoi_list,
  sig_color = "#F39B7FB2",
  non_sig_color = "gray"
)

Arguments

tkoi_list

A tKOIList object returned by run_tkoi(). It uses the expression_data slot (columns gene_name, logfc, and pvalue), the Gene table of network_summary_statistics (columns node_id, name, and fdr), and the pvalue_threshold and logfc_threshold of the run.

sig_color

Fill color of genes whose abs(logfc) is at least the run's logfc_threshold. Default "#F39B7FB2".

non_sig_color

Fill color of the other genes. Default "gray".

Value

A ggplot object.

Details

expression_data is cleaned the same way as in run_tkoi(): rows with a missing or blank gene_name are dropped, and only the first row of each gene is used. Genes are matched to network nodes through their Ensembl IDs (see genes); network genes without expression data are not plotted.

In the plot:

• The x-axis is -log10(pvalue) (experimental p-value).

• The y-axis is -log10(fdr) (tKOI FDR).

• Genes with logfc >= 0 are in the Up-regulated panel, the others in the Down-regulated panel.

• The dashed vertical line marks the run's pvalue_threshold and the dashed horizontal line an FDR of 0.05.

See also

export_gene_exploration_data() for the underlying table.

Examples

if (FALSE) { # \dontrun{
expression_data = data.table::fread(
  system.file("extdata", "example_data.csv", package = "tkoi")
)

set.seed(1)
tkoi_result = run_tkoi(expression_data = expression_data)

plt = make_gene_exploration_plot(tkoi_result, sig_color = "#F39B7FB2", non_sig_color = "gray")
plt
} # }