logistic 模型构建

使用的是R包glmnet 函数，有些时候由于方法不收敛可以增加迭代次数，使得模型收敛。

log_res<-glm(pheno~PRS,family=binomial(link='logit'),control=list(maxit=150),data=train_data)

summary(log_res)

logistic.display(log_res)

交叉验证：1000次10倍交叉验证

install.packages('caret')

install.packages('lattice')

library('caret')

set.seed(7)

require("RColorBrewer")

require(caret)

colss=brewer.pal(8,"Set1")[1:10]

display.brewer.all(type = "qual")

for(j in 1:1000){

pdf(paste0(j,"_perm.pdf",sep=""),width=6,height=6)

  a<-dev.cur()

  png(paste0(j,"_perm.png",sep=""))

  dev.control("enable")

 folds<-createFolds(y=train_data$pheno,k=5)

  for (i in 1:5)

  {

    print(i)

   fold_test<-train_data[folds[[i]],]

    fold_train<-train_data[-folds[[i]],]

    print ("进行中")

    ##fold_pre<-glm(pheno~age+gender+PRS,family=binomial(link='logit'),data=fold_train)

    ##fold_pre<-glm(pheno~PRS,family=binomial(link='logit'),data=fold_train)

    summary(fold_pre)

    fold_predict <- predict.glm(fold_pre,type='response',newdata=fold_test)

    fold_predicts = ifelse(fold_predict > 0.5,1,0)

    fold_test$predict = fold_predicts

    fold_real<-fold_test$pheno

    modelroc<-roc(fold_real,fold_predict)

    print(modelroc$auc)

    if(i==1)

    {

      #plot.roc(modelroc, percent = TRUE, main="Smoothing",print.auc =TRUE)

      plot(modelroc,col=colss[i],

          auc.polygon = TRUE,max.auc.polygon=TRUE,

          max.auc.polygon.col="white",auc.polygon.col = "white",

          legacy.axes=TRUE,lwd=2,identity=F,main="ROC curve")

    }

    else

    {

      plot(modelroc, add=TRUE,col=colss[i],

          max.auc.polygon.col="white",

          auc.polygon.col = "white",legacy.axes=TRUE,lwd=2,identity=F)

    }

    legend(0.5,0.1+0.05*i,paste("AUC=",signif(modelroc$auc,3)),col=colss[i],lty=2,cex=1.5,bty="n")

  }

  dev.copy(which=a)

  dev.off()

  dev.off()

}

```