Skip to contents

Export Gene Exploration Data

Source: R/export_gene_exploration_data.R
export_gene_exploration_data.Rd

Combines the differential expression table of a run_tkoi() result with the tKOI network statistics of every gene node, so that experimental and network evidence can be compared gene by gene.

Usage

export_gene_exploration_data(tkoi_list)

Arguments

tkoi_list

A tKOIList object returned by run_tkoi(). Two slots are used:

  • expression_data: the input table, with columns gene_name (Ensembl gene IDs), logfc, and pvalue.

  • network_summary_statistics: its Gene table, with columns node_id, identifier, name, pagerank, beta, p_value, and fdr.

Value

A data frame with one row per row of the Gene table and columns:

  • gene_name: Ensembl gene ID from expression_data.

  • gene_symbol: Gene symbol.

  • id: Node ID of the gene in the network (node_id in the Gene table).

  • identifier: NCBI Entrez Gene ID, as a character string.

  • experimental_logfc: logfc from expression_data.

  • experimental_pvalue: pvalue from expression_data.

  • pagerank: Observed personalized PageRank.

  • tkoi_beta: tKOI network enrichment z-score (beta).

  • tkoi_pvalue: Unadjusted one-sided tKOI p-value.

  • tkoi_fdr: tKOI false discovery rate (Benjamini-Hochberg, among gene nodes).

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). Every row of the Gene table is kept, so network genes that are not in expression_data have NA in gene_name, experimental_logfc, and experimental_pvalue.

See also

make_gene_exploration_plot() to plot the same data, export_network_summary_statistics() to export every node type.

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)

gene_data = export_gene_exploration_data(tkoi_result)
head(gene_data)

# Genes supported by both the experiment and the network
subset(gene_data, experimental_pvalue <= 0.05 & tkoi_fdr <= 0.05)
} # }