createGeneModel: Create a model for a target gene using a Trena object

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

Description

Create a model for a target gene using a Trena object

Usage

## S4 method for signature 'Trena'
createGeneModel(obj, targetGene, solverNames,
  tbl.regulatoryRegions, mtx)

Arguments

obj	An object of class Trena
targetGene	The name of a target gene to use for building a model
solverNames	A character vector containing the solver names to be used for building the model
tbl.regulatoryRegions	A data frame of regulatory regions, typically generated by using a filter
mtx	An assay matrix of expression data

Value

A data frame containing the gene model

Examples

# Create a Trena object for human and make a gene model for "MEF2C" using a footprint filter
trena <- Trena("hg38")
chromosome <- "chr5"
mef2c.tss <- 88904257
loc.start <- mef2c.tss - 1000
loc.end   <- mef2c.tss + 1000

database.filename <- system.file(package="trena", "extdata", "mef2c.neigborhood.hg38.footprints.db")
database.uri <- sprintf("sqlite://%s", database.filename)
sources <- c(database.uri)
load(system.file(package="trena", "extdata/ampAD.154genes.mef2cTFs.278samples.RData"))

motifs.list <- getRegulatoryChromosomalRegions(trena, chromosome, mef2c.tss-1000, mef2c.tss+1000,
sources, "MEF2C", mef2c.tss)

library(MotifDb)
tbl.motifs.tfs <- associateTranscriptionFactors(MotifDb, motifs.list[[1]], source="MotifDb", expand.rows=TRUE)
model.mef2c <- createGeneModel(trena, "MEF2C", c("lasso","ridge","randomforest"), tbl.motifs.tfs, mtx.sub)

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