solve.RandomForest: Run the Random Forest Solver

In PriceLab/trena-until-01mar2018: Fit transcriptional regulatory networks using gene expression, priors, machine learning

Description

Given a TReNA object with RandomForest as the solver, use the randomForest function to estimate coefficients for each transcription factor as a predictor of the target gene's expression level.

Usage

## S4 method for signature 'RandomForestSolver'
run(obj)

Arguments

obj	An object of class TReNA with "randomForest" as the solver string

Value

A data frame containing the IncNodePurity for each candidate regulator. This coefficient estimates the relationship between the candidates and the target gene.

See Also

randomForest, RandomForestSolver

Other solver methods: run,BayesSpikeSolver-method, run,EnsembleSolver-method, run,LassoPVSolver-method, run,LassoSolver-method, run,PearsonSolver-method, run,RidgeSolver-method, run,SpearmanSolver-method, run,SqrtLassoSolver-method

Examples

# Load included Alzheimer's data, create a TReNA object with Random Forest as solver, and solve
load(system.file(package="trena", "extdata/ampAD.154genes.mef2cTFs.278samples.RData"))
targetGene <- "MEF2C"
candidateRegulators <- setdiff(rownames(mtx.sub), targetGene)
rf.solver <- RandomForestSolver(mtx.sub, targetGene, candidateRegulators)
tbl <- run(rf.solver)

PriceLab/trena-until-01mar2018 documentation built on May 25, 2019, 1:22 p.m.