modelGeneExpression: The modelGeneExpression function

In mpru/BIMEGA: BIMEGA: BIvariate Methylation and Expression GAussian mixture model

Description

Model gene expression as a function of gene expression with a polynomial robust regression model (robust base package). Up to grade 3 models are evaluated and compared. R-squared and estimated coefficients are returned for the best model only for the genes with significant models.

Usage

modelGeneExpression(METcancer, MAcancer, CovariateData = NULL)

Arguments

METcancer	matrix with methylation data for cancer samples (genes in rows, samples in columns).
MAcancer	matrix with gene expression data for cancer samples (genes in rows, samples in columns).
CovariateData	vector (numeric or character) indicating a covariate to be included in the model to adjust for it. Not used in an standard run of BIMEGA. It can be used if samples can from different tissue type, for example.

Value

matrix with R-square, degree of the best polynomial model and estimated coefficients only for the significant genes.

Examples

# load data sets
data(METcancer)
data(MAcancer)

# model gene expression
results <- modelGeneExpression(METcancer, MAcancer)

mpru/BIMEGA documentation built on May 23, 2019, 6:34 a.m.