findDampeningConstant: findDampeningConstant

In DWLS: Gene Expression Deconvolution Using Dampened Weighted Least Squares

findDampeningConstant

Description

Finds a dampening constant for the weights using cross-validation. The goldStandard is used to define the weights. Multiple values of the dampening constant (multiplier) are tried. For each attempt, the variance of the dampened weighted solution for a subset of genes is calculated (on a randomly selected half of the genes). Note that infinite weights are ignored.The dampening constant that results in least cross-validation variance is chosen. It functions in a nondeterministic manner. The dampening constant defines the maximum value that any weight can take on.

Usage

findDampeningConstant(S, B, goldStandard)

Arguments

S	List output from trimData$Sig (S)
B	List output from trimData$dataBulk (B)
goldStandard	Starting point for the weights, determined by solving OLS

Value

value (dampening constant value)

Examples

#Sig
#url <- "https://github.com/sistia01/DWLS/raw/main/inst/extdata/Sig.RData"
#dest <- "data/Sig.RData"
#download.file(url, dest)
#load("data/Sig.RData")
load(system.file("extdata", "Sig.RData", package = "DWLS"))

#dataBulk
#url <- "https://github.com/sistia01/DWLS/raw/main/inst/extdata/dataBulk.RData"
#dest <- "data/dataBulk.RData"
#download.file(url, dest)
#load("data/dataBulk.RData")
load(system.file("extdata", "dataBulk.RData", package = "DWLS"))

trimmed <- trimData(Sig, dataBulk)
S <- trimmed$sig
B <- trimmed$bulk
solution <- solveOLSInternal(S,B)
findDampeningConstant(S, B, solution)

DWLS documentation built on May 24, 2022, 5:07 p.m.