#random forest
library(randomForest)

# Read in data for RF classifier with SIFT, GO, and PfamLogR.E-value
trainingData.Sift.Go.Pfam = read.table(file="training.data.sift.go.pfam",sep="\t",header=T)

# Train RF classifier
forestSGP = randomForest(Class ~ Sift + Pfam + Go,data=trainingData.Sift.Go.Pfam,importance=T,keep.forest=T, ntree=1000, mtry=2)  

# Load test data
dataTest.sgp = read.table("test.data.crossValidation.sift.go.pfam",sep="\t",header=T)

# Classify variants
test.forest.sgp = predict(forestSGP,dataTest.sgp[3:5])