CombineDependencies: Combine dCor output

In tallulandrews/scTarNet: Single-cell Targetted Network Inference

Description

Combines output from calculateTFstoTargets for multiple batches/normalizations.

Usage

	combineDependencies(list_of_correlation_tables)

Arguments

list_of_correlation_tables

list of output from calculateTFstoTargets

Details

Counts the number of times TF-Target inferred across multiple batches or normalizations. And infers the consensus direction of the relationship as follows: if always "other" (0) then "other" (0) if always "other" (0) or "positive" (1) then positive (1) if always "other" (0) or "negative" (1) then negative (1) if mix of "positive" and "negative" then "other" (0)

Value

A table with columns "Gene" (TF), "Target", "recurr" (number of batches), and "direction"

Examples

	set.seed(13)
	batchA_output <- data.frame(
		Gene=sample(c("A", "B", "C", "D"), 50, replace=TRUE),
		Target=sample(c("E", "F", "G", "E", "F"), 50, replace=TRUE)
		)
	batchA_output <- unique(batchA_output)
	batchA_output$estimate <- rep(1, times=nrow(batchA_output))
	batchA_output$pval <- rep(1, times=nrow(batchA_output))
	batchA_output$direction <- sample(c(1,0,-1), nrow(batchA_output), replace=TRUE)


	batchB_output <- data.frame(
		Gene=sample(c("A", "B", "C", "D"), 50, replace=TRUE),
		Target=sample(c("E", "F", "G", "E", "F"), 50, replace=TRUE)
		)
	batchB_output <- unique(batchB_output)
	batchB_output$estimate <- rep(1, times=nrow(batchB_output))
	batchB_output$pval <- rep(1, times=nrow(batchB_output))
	batchB_output$direction <- sample(c(1,0,-1), nrow(batchB_output), replace=TRUE)

	consensus <- combineDependencies(list(batchA_output, batchB_output));

tallulandrews/scTarNet documentation built on May 31, 2019, 2:27 p.m.