Skip to contents

Generate reciprocal best hits based on set similarities

rs_rbh_sets.Rd

[Experimental] This function takes a nested list that contains gene modules/sets derived from various methods and identifies (k-th) reciprocal best hits between gene modules/sets across every pair of origins.

Usage

rs_rbh_sets(module_list, k_best, overlap_coefficient, min_similarity)

Arguments

module_list

A nested named list. The outer list should contain the origin of the gene modules, the inner list the names of the gene modules and the respective genes in them.

k_best

Integer. Number of best neighbours to consider. If set to 1L, this behaves as the traditional reciprocal best hit. If you set this to 3L you consider edges if the modules is in the top 3 best modules by similarity for each other.

overlap_coefficient

Boolean. Shall the overlap coefficient instead of the Jaccard similarity be used.

min_similarity

Numeric. Only hits with a similarity strictly above this are returned.

Value

A list containing:

  • origin - The name of the origin of the gene modules.

  • target - The name of the target of the gene modules.

  • comparisons - Integer vector indicating how many RBH hits were identified in this comparison

  • origin_modules - Names of the gene modules from the origin.

  • target_modules - Names of the gene modules from the target.

  • similarity - The similarities between the two respective gene modules.