R语言-专题

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

### 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)  #返回矩阵

### 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") #只替换第一个
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
mutate(test, new = Sepal.Length * Sepal.Width) #new为新增列的名字

# 1.读取group.csv,从第二列中提取圈出来的信息
library(stringr)
a = read.csv("group.csv")
g  = str_split(a$title," ",simplify = T)
g
g[,4]
# 2.如何把上一题结果中的Control和Vemurafenib改成全部小写？搜索一下
tolower(g[,4])
str_to_lower(g[,4])

#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")  #警惕写的代码结果顺序和实际相反

#### (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"))

library(ggplot2)
a = data.frame(a1 = rnorm(15),
               a2 = rnorm(15),
               a3 = rnorm(15),
               g = rep(c("a","b","c"),each = 5))
a

#下标循环不保存
library(ggplot2)
for (i in 1:3) {
  p = ggplot(data = a)+
    geom_boxplot(aes_string(y = a1[[i]],
                     x = g))
  print(p)
}

# 元素循环，不方便保存
for (i in colnames(a)[1:3]) {
 p =  ggplot(data = a)+
    geom_boxplot(aes_string(y = i,
                     x = "g"))

  print(p) 
}

# 下标循环，可以保存数据
p = list()
for (i in 1:3) {
  p[[i]] =  ggplot(data = a)+
    geom_boxplot(aes_string(y = colnames(a)[[i]],
                            x = "g"))
  
  print(p[[i]]) 
}
library(patchwork)
p[[1]]+p[[2]]+p[[3]]

# 表达矩阵
set.seed(10086)
exp = matrix(rnorm(18),ncol = 6)
exp = round(exp,2)
rownames(exp) = paste0("gene",1:3)
colnames(exp) = paste0("test",1:6)
exp[,1:3] = exp[,1:3]+1
exp

library(tidyr)
library(tibble)
library(dplyr)
dat = t(exp) %>%  #转置
  as.data.frame() %>% 
  rownames_to_column() %>%  #rownames变成一列
  mutate(group = rep(c("control","treat"),each = 3)) #新增一列

pdat = dat%>%  #宽变长
  pivot_longer(cols = starts_with("gene"),
               names_to = "gene",
               values_to = "count")

library(ggplot2)
p = ggplot(pdat,aes(gene,count))+
  geom_boxplot(aes(fill = group))+
  theme_bw()
p
p + facet_wrap(~gene,scales = "free") #分面

a= norm(100)
tail(sort(a),10)

### 2.lapply(list, FUN, …) 
# 对列表/向量中的每个元素（向量）实施相同的操作

test <- list(x = 36:33,y = 32:35,z = 30:27);test

#返回值是列表，对列表中的每个元素（向量）求均值(试试方差var,分位数quantile)

lapply(test,mean)
lapply(test,fivenum)

### 3.sapply 简化结果，返回矩阵或向量

sapply(test,mean)
sapply(test,fivenum)

class(sapply(test,fivenum))