calculateThreshold: Calculate the default threshold of edge-weight in PANDA...

In twangxxx/netZoo: Integrate methods: PANDA, LIONESS, CONDOR, ALPACA into one workflow.

Description

In a PANDA network, the 4th column is filled with the edge-weight calculated by Z-score formular. [(Glass et al. 2013)]) This function transforms all the edge-weights into values above 0 by formular w'=ln(e^w+1) , then calculates the midpoint between the median edge-weight of prior ( 3rd column "Motif" is 1.0) edges and the median edge-weight of non-prior edges (3rd column "Motif" is 0.0) in PANDA network.

Usage

calculateThreshold(df)

Arguments

df	Data Frame indicating the entire network result of PANDA algorithm, it is created by runPanda and accessd by $panda.

Value

Numeric vector of the midpoint between two medians. One median is the median edge-weight of all prior edges, another is the meadian edge-weight of all non-prior edges.

Examples

# refer to four input datasets files in inst/extdat
treated_expression_file_path <- system.file("extdata", "expr4.txt", package = "netZoo", mustWork = TRUE)
control_expression_file_path <- system.file("extdata", "expr10.txt", package = "netZoo", mustWork = TRUE)
motif_file_path <- system.file("extdata", "chip.txt", package = "netZoo", mustWork = TRUE)
ppi_file_path <- system.file("extdata", "ppi.txt", package = "netZoo", mustWork = TRUE)


# Run PANDA for treated and control network
treated_all_panda_result <- runPanda(e = treated_expression_file_path, m = motif_file_path, ppi = ppi_file_path, rm_missing = TRUE )
control_all_panda_result <- runPanda(e = control_expression_file_path, m = motif_file_path, ppi = ppi_file_path, rm_missing = TRUE )

# Access PANDA regulatory network
treated_net <- treated_all_panda_result$panda
control_net <- control_all_panda_result$panda

# Calculate the default threshold of edge-weight in a PANDA network.
calculateCutoff (treated_net )

twangxxx/netZoo documentation built on May 17, 2019, 1:02 p.m.