library(glmnet)
setwd("") 
train=read.table("train.txt",header=T,sep="\t",check.names=F,row.names=1) 
test=read.table("test.txt",header=T,sep="\t",check.names=F,row.names=1)            
test=test[,colnames(train)]
rt=rbind(train,test)

x=as.matrix(train[,c(2:ncol(train))])
y=train[,1]
fit=glmnet(x, y, family = "binomial", alpha=1)
cvfit=cv.glmnet(x, y, family="binomial", alpha=1,type.measure='auc',nfolds =10)
pdf(file="cvfit.pdf",width=6,height=5.5)
plot(cvfit)
dev.off()

coef=coef(fit, s = cvfit$lambda.min)
index=which(coef != 0)
actCoef=coef[index]
lassoGene=row.names(coef)[index]
coef=cbind(Gene=lassoGene,Coef=actCoef)
write.table(coef,file="coef.txt",sep="\t",quote=F,row.names=F)
write.table(lassoGene[-1],file="miRNAlist.txt",sep="\t",quote=F,row.names=F,col.names=F)

threshold=0.5
bioStat=function(x=null, y=null, dataType=null){
  pred=predict(cvfit,newx=x,s=cvfit$lambda.min,type = 'response')
  outTab=cbind(id=row.names(pred),TMB=as.character(y),pred)
  colnames(outTab)=c("id","MDR","value")
  sameGene=intersect(colnames(x),lassoGene)
  outTab=cbind(outTab,x[,sameGene])
  write.table(outTab,file=paste0("value.",dataType,".txt"),sep="\t",quote=F,row.names=F)
  
  pred_new=as.integer(pred>threshold) 
  tab=table(pred_new,y)
  TP=tab[2,2];TN=tab[1,1];FP=tab[2,1];FN=tab[1,2]
  Accuracy=round((TP+TN)/(TP+FN+FP+TN),4)
  SE=round(TP/(TP+FN),4)
  SP=round(TN/(TN+FP),4)
  PPV=round(TP/(TP+FP),4)
  NPV=round(TN/(TN+FN),4)
  

  rocobj1=roc(y, as.vector(pred))
  AUC=auc(rocobj1)
  AUC=round(AUC,4)
  return(c(SE,SP,PPV,NPV,Accuracy,AUC))
}


trainValue=bioStat(x=x,y=y,dataType="train")

testX=as.matrix(test[,c(2:ncol(test))])
testY=test[,1]
testValue=bioStat(x=testX,y=testY,dataType="test")

totalX=as.matrix(rt[,c(2:ncol(rt))])
totalY=rt[,1]
totalValue=bioStat(x=totalX,y=totalY,dataType="all")


statTab=rbind(Train=trainValue,Test=testValue,Total=totalValue)
colnames(statTab)=c("SE","SP","PPV","NPV","Accuracy","AUC")
statTab=rbind(ID=colnames(statTab),statTab)
write.table(statTab,file="accuracyStat.xls",sep="\t",col.names=F,quote=F)