R语言基础：环境搭建、常用代码

1. java环境配置

2. R包安装

#方法一
source("http://bioconductor.org/biocLite.R")
biocLite("kernlab")

#方法二
install.packages("kernlab")

#方法三
if(!suppressWarnings(require('kernlab')))
{
  install.packages('kernlab')
  require('kernlab')
}

#方法四（推荐）
library(BiocManager)
BiocManager::install("plot3D",ask = F,update = F)

3. R包卸载/卸除/查看加载

#卸载
remove.packages("rlang")
remove. packages(c("pkg1","pkg2") , lib = file .path("path", "to", "library"))

#卸除：在包使用函数冲突，检验函数依赖时比较有用
detach("package:RMySQL")

#已加载
(.packages())

4. Rstudio设置

（1）ctrl+Alt+Shift+0 显示四个窗口

（2）R版本共存

• https://segmentfault.com/a/1190000015332215?utm_source=tag-newest
• https://blog.csdn.net/ChenQihome9/article/details/81949965
• https://cran.r-project.org/bin/windows/base/old/

5. 文件读取

• （1）读写excel

library(readxl)
data<-read_xlsx('E:/代码/机器学习/winequality-red.xlsx', sheet = 'Sheet1')

library(openxlsx)
data<-read.xlsx('E:/代码/机器学习/winequality-red.xlsx')

#install.packages("rJava")
#install.packages("xlsx")
#Sys.setenv(JAVA_HOME='C:/Program Files/Java/jdk1.8.0_171/jre')
library(rJava)
library(xlsxjars)
library(xlsx)
data<-read.xlsx(workbook, 1, encoding='UTF-8')  #name sheet 防乱码
write.xlsx(BP, paste('GO_BP_Module_', i, '.xlsx', sep=""), row.names = F)
write.xlsx(BP, 'target_GO_BP.xlsx') 

• （2）读写txt

data<-read.table(file.choose(),header=T, sep="\t")

6. 常用数据处理函数/代码

• 参考：https://www.cnblogs.com/zongfa/p/8537031.html （dplyr、tidyr包）

• 多向量取交集

    A<-c()
    B<-c()
    C<-c()
    result<-Reduce(intersect, list(A, B, C))

• 横向输出

    library(tidyr)
    data<-as.data.frame(t(data)) #任意向量
    colnames(data)<-c(1:dim(data)[2])
    result<-unite(data,"combine",colnames(data), sep="\t", remove = F)[,1] #combine可改，、\t空格间隔

• 拆分填充行

    #excel 初步处理得到互作对
    ncRDeath<- read.table(file.choose(),sep = "\t", stringsAsFactors = F,header = T, fill=TRUE, quote = "")
    head(ncRDeath)
    ncRDeath2<-ncRDeath[ncRDeath$Gene_Symbol != "" &      ncRDeath$miRBase_mature_ID != "",]

    final<-data.frame()
    for (i in 1:dim(ncRDeath2)[1]){
     #i=1
     ncR_break<-data.frame(miRBase_mature_ID=unlist(strsplit(ncRDeath2[i,1], ",")),Gene_Symbol=ncRDeath2[i,2])
     final<-rbind(final,ncR_break)
     print(i)
    }
    dim(ncRDeath2) #2119
    write.table(final,"6-ncRdeathDB_miRNA_break.txt",row.names = F,quote = F,sep="\t")

• 重置和变换

    #（1）重置变量名
    library(reshape)
    rename(ncrdeathDB_miRNA, miRBase_mature_ID="miRNA", Gene_Symbol="target_mRNA")

    #（2）重置数据框
    inputData<-data.frame(cats[, c (2,3)], response = as.factor(cats$Sex))

    #（3）正确添加表头：将行名加入矩阵
    datamatrix= cbind(rownames(datamatrix),datamatrix)
    rownames(datamatrix)[1]<-'A'

7. 清除所有变量

rm(list = ls())

8. 计算运行时间

exeTimeTsne<-system.time(tsne)

9. 平台对接

• （1）与SPSS对接

setwd('')
#install.packages('foreign')
library(foreign)
data.spss<-read.spss("car_sales.sav", to.data.frame = TRUE)
head(data.spss)

#install.packages('Hmisc')
library(Hmisc)
mydataframe<-spss.get("C:/Users/Administrator/Desktop/test.sav", use.value.lables=TRUE)

• （2）与python对接（不成功）
  参考：
• https://www.jianshu.com/p/06dbd8dcc198
• https://stackoverflow.com/questions/59321283/reticulate-py-available-returns-false

#install.packages("reticulate")
library(reticulate)