RFpurify

R-package to predict ABSOLUTE and ESTIMATE tumor purity using a Random Forest regression model trained on TCGA Illumina 450k DNA-methylation array data. Purity prediction functions can be applied to 450k and 850k array data preprocessed with the Bioconductor package minfi.

Install minfi and RF_purify package from GitHub

if(!require(minfi)){
  if(!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
  BiocManager::install("minfi")
}

if(!require(devtools)) install.packages("devtools")   
library(devtools)
# prevent warnings from beeing converted to errors when calling install_github
Sys.setenv(R_REMOTES_NO_ERRORS_FROM_WARNINGS=TRUE)
install_github('mwsill/RFpurify')

Example

Install minfiData with example data

if(!require(minfiData)){
  if(!requireNamespace("BiocManager", quietly = TRUE))
    install.packages("BiocManager")
  BiocManager::install("minfiData")
}

Predict tumor purity

library(RFpurify)
library(minfiData)

# load example data
data(MsetEx)
MsetEx

## class: MethylSet 
## dim: 485512 6 
## metadata(0):
## assays(2): Meth Unmeth
## rownames(485512): cg00050873 cg00212031 ... ch.22.47579720R
##   ch.22.48274842R
## rowData names(0):
## colnames(6): 5723646052_R02C02 5723646052_R04C01 ...
##   5723646053_R05C02 5723646053_R06C02
## colData names(13): Sample_Name Sample_Well ... Basename filenames
## Annotation
##   array: IlluminaHumanMethylation450k
##   annotation: ilmn12.hg19
## Preprocessing
##   Method: Raw (no normalization or bg correction)
##   minfi version: 1.21.2
##   Manifest version: 0.4.0

# predict purity
absolute <- predict_purity(MsetEx,method="ABSOLUTE")
estimate <- predict_purity(MsetEx,method="ESTIMATE")

# plot predicted ABSOLUTE purity against ESTIMATE purity
plot(x=estimate,y=absolute,pch=19,ylim=c(0,1),xlim=c(0,1))
abline(a=0,b=1,col="red")