optimize_numRegions: Optimize number of regions used for prediction

In leekgroup/phenopredict: Predicting phenotypes from RNA-seq expression data

Description

This function takes a list of possible options for the numRegions argument of build_predictor. Using this set of possible numRegions and expression data (the training data / output of filter_regions()), this function builds a predictor for each possible numRegions. Prediction accuracy is then calculated across varying numbers of regions. The numRegions argument that optimizes accuracy in the training data can then be used in build_predictor.

Usage

optimize_numRegions(inputdata = NULL, phenodata = NULL, phenotype = NULL,
  covariates = NULL, type = NULL, numRegions_set = c(10, 20, 30, 60, 80,
  100, 150, 200))

Arguments

inputdata	output from filter_regions() inputdata
phenodata	data set with phenotype information; samples in rows, variables in columns phenodata
phenotype	phenotype of interest phenotype
covariates	Which covariates to include in model covariates
type	The class of the phenotype of interest (numeric, binary, factor) type
numRegions_set	set of numRegions to test numRegions_set

Value

Prediction accuracies across each numRegions argument tested

Examples

library('GenomicRanges')
library('dplyr')

## Make up some some region data
regions <- GRanges(seqnames = 'chr2', IRanges(
     start = c(28971710:28971712, 29555081:29555083, 29754982:29754984),
     end = c(29462417:29462419, 29923338:29923340, 29917714:29917716)))

## make up some expression data for 9 rows and 30 people
data(sysdata, package='phenopredict')
## includes R object 'cm'
exp= cm[1:length(regions),1:30]

## generate some phenotype information
sex = as.data.frame(rep(c("male","female"),each=15))
age = as.data.frame(sample(1:100,30))
pheno = dplyr::bind_cols(sex,age)
colnames(pheno) <- c("sex","age")

## filter regions to be used to build the predictor
inputdata <- filter_regions(expression=exp, regiondata=regions,
	phenodata=pheno, phenotype="sex", covariates=NULL,
	type="factor", numRegions=5)

regnum <- optimize_numRegions(inputdata=inputdata ,phenodata=pheno,
phenotype="sex", covariates=NULL,type="factor",numRegions_set=c(3,5))

leekgroup/phenopredict documentation built on May 14, 2019, 11:27 a.m.