comp_fea_res: Plot for Comparing Ranking Results of FEA Methods

In signatureSearch: Environment for Gene Expression Searching Combined with Functional Enrichment Analysis

Description

Dot plot for comparing the top ranking functional categories from different functional enrichment analysis (FEA) results. The functional categories are plotted in the order defined by their mean rank across the corresponding FEA results.

Usage

comp_fea_res(
  table_list,
  rank_stat = "pvalue",
  Nshow = 20,
  Nchar = 50,
  scien = FALSE,
  ...
)

Arguments

table_list	a named list of tibbles extracted from feaResult objects, e.g. generated with different FEA methods.
rank_stat	character(1), column name of the enrichment statisic used for ranking the functional categories, e.g. 'pvalue' or 'p.adjust'. Note, the chosen column name needs to be present in each tibble of 'table_list'.
Nshow	integer defining the number of the top functional categories to display in the plot after re-ranking them across FEA methods
Nchar	integer defining number of characters displayed (exceeded characters were replaced by '...') in the description of each item
scien	TRUE or FALSE, indicating whether the rank_stat is rounded to the scientific format with 3 digits
...	Other arguments passed on to geom_point

Details

The 'comp_fea_res' function computes the mean rank for each functional category across different FEA result instances and then re-ranks them based on that. Since the functional categories are not always present in all enrichment results, the mean rank of a functional category is corrected by an adjustment factor that is the number of enrichment result methods used divided by the number of occurences of a functional category. For instance, if a functional category is only present in the result of one method, its mean rank will be increased accordingly. Subsequently, the re-ranked functional categories are compared in a dot plot where the colors represent the values of the enrichment statistic chosen under the rank_stat argument.

Value

ggplot2 graphics object

Examples

method1 <- data.frame("ID"=paste0("GO:", 1:5), 
                      "Description"=paste0("desc", 1:5),
                      "pvalue"=c(0.0001, 0.002, 0.004, 0.01, 0.05))
method2 <- data.frame("ID"=paste0("GO:", c(1,3,5,4,6)), 
                      "Description"=paste0("desc", c(1,3,5,4,6)),
                      "pvalue"=c(0.0003, 0.0007, 0.003, 0.006, 0.04))
table_list <- list("method1" = method1, "method2"=method2) 
comp_fea_res(table_list, rank_stat="pvalue", Nshow=20)

signatureSearch documentation built on April 16, 2021, 6 p.m.