阿榜的生信笔记7—R语言的综合运用1

rm(list = ls())
if(!require(stringr))install.packages('stringr')
library(stringr)

x <- "The birch canoe slid on the smooth planks."
x
### 1.检测字符串长度
str_length(x)
length(x)

### 2.字符串拆分
str_split(x," ")
x2 = str_split(x," ")[[1]];x2

y = c("jimmy 150","nicker 140","tony 152")
str_split(y," ")
str_split(y," ",simplify = T)

### 3.按位置提取字符串
str_sub(x,5,9)

### 4.字符检测
str_detect(x2,"h")
str_starts(x2,"T")
str_ends(x2,"e")

### 5.字符串替换
x2
str_replace(x2,"o","A")
str_replace_all(x2,"o","A")

### 6.字符删除
x
str_remove(x," ")
str_remove_all(x," ")

test <- iris[c(1:2,51:52,101:102),]
rownames(test) =NULL # 去掉行名，NULL是“什么都没有”
test

# arrange，数据框按照某一列排序

library(dplyr)
arrange(test, Sepal.Length) #从小到大
arrange(test, desc(Sepal.Length)) #从大到小

# distinct，数据框按照某一列去重复
distinct(test,Species,.keep_all = T)

# mutate，数据框新增一列
test=mutate(test, new = Sepal.Length * Sepal.Width)

# 连续的步骤

# 1.多次赋值，产生多个中间的变量
x1 = select(iris,-5)
x2 = as.matrix(x1)
x3 = head(x2,50)
pheatmap::pheatmap(x3)

# 2. 嵌套，代码不易读
pheatmap::pheatmap(head(as.matrix(select(iris,-5)),50))

# 3.管道符号传递，简洁明了
iris %>%
  select(-5) %>%
  as.matrix() %>%
  head(50) %>% 
  pheatmap::pheatmap()

rm(list = ls())

## 一.条件语句

###1.if(){ }

#### (1)只有if没有else，那么条件是FALSE时就什么都不做

i = -1
if (i<0) print('up')
if (i>0) print('up')

#理解下面代码
if(!require(tidyr)) install.packages('tidyr')

#### (2)有else
i =1
if (i>0){
  print('+')
} else {
  print("-")
}
i = 1
ifelse(i>0,"+","-")

x = rnorm(3)
x
ifelse(x>0,"+","-")

#ifelse()+str_detect(),王炸
samples = c("tumor1","tumor2","tumor3","normal1","normal2","normal3")
k1 = str_detect(samples,"tumor");k1
ifelse(k1,"tumor","normal")
k2 = str_detect(samples,"normal");k2
ifelse(k2,"normal","tumor")

#### (3)多个条件
i = 0
if (i>0){
  print('+')
} else if (i==0) {
  print('0')
} else if (i< 0){
  print('-')
}

ifelse(i>0,"+",ifelse(i<0,"-","0"))

## 二、for循环

for( i in 1:4){
  print(paste0("the current number is ",i))
}

#批量画图
par(mfrow = c(2,2))
for(i in 1:4){
  print(paste0("the current column is ",colnames(iris)[i]))
  plot(iris[,i],col = iris[,5])
}
#批量装包
pks = c("tidyr","dplyr","stringr")
for(g in pks){
  print(paste0("package ",g," is installing"))
  if(!require(g,character.only = T))
    install.packages(g,ask = F,update = F)
}