一、单个数据框处理⭐️⭐️⭐️——使用前加载dplyr

1、sort(),order()排序书写太麻烦？试试arrange()吧⭐️⭐️⭐️

> library(dplyr)
> arrange(test, Sepal.Length) #按照Sepal.Length列升序排列
> arrange(test, desc(Sepal.Length))
> arrange(test, desc(Sepal.Width),Sepal.Length)#双排序（若A列相同，再按B列排）
Sepal.Length Sepal.Width Petal.Length Petal.Width    Species
1          5.1         3.5          1.4         0.2     setosa
2          6.3         3.3          6.0         2.5  virginica
3          6.4         3.2          4.5         1.5 versicolor
4          7.0         3.2          4.7         1.4 versicolor
5          4.9         3.0          1.4         0.2     setosa
6          5.8         2.7          5.1         1.9  virginica

2、新增列——mutate()

test=mutate(test,new=Sepal.Length*Sepal.Width)

3、筛选行、列——filter(),select()

4、管道符号——避免中间变量过多（%>%),理解为”then”

二、将表达矩阵绘制为箱图⭐️⭐️⭐️⭐️（需将数据预处理）

——————>(宽变长）

set.seed(10086)
exp = matrix(rnorm(18),ncol = 6)
exp = round(exp,2)#保留两位小数
rownames(exp) = paste0("gene",1:3)
colnames(exp) = paste0("test",1:6)
exp[,1:3] = exp[,1:3]+1 #人工处理使差异明显
exp
library(tidyr)
library(tibble)
library(dplyr)
dat = t(exp) %>% 
  as.data.frame() %>% 
  rownames_to_column() %>% #行名变列
  mutate(group = rep(c("control","treat"),each = 3))
pdat = dat%>% 
  pivot_longer(cols = starts_with("gene"),#合并哪些列
               names_to = "gene",#给新列列名
               values_to = "count")#未处理的值放入此列
library(ggplot2)
p = ggplot(pdat,aes(gene,count))+
  geom_boxplot(aes(fill = group))+
  geom_jitter()+
  theme_bw()
p
p + facet_wrap(~gene,scales = "free") #scales = "free"是作图时x,y轴是否要按比例（默认按比例）

三、缺失值处理⭐️⭐️⭐️——个人认为重要，测序结果有时遇到缺失

> drop_na(X) #把所有有Na值行删除
> drop_na(X,X1)#查看特定列缺失值，并去除该行
#替换NA
> replace_na(X,list(X2=0))
#用上一行的值填充NA
fill(X,X2) #若NA为第一个值则仍为NA

四、多个数据框处理

> test1 <- data.frame(name = c('jimmy','nicker','Damon','Sophie'), 
+                     blood_type = c("A","B","O","AB"))
> test1
    name blood_type
1  jimmy          A
2 nicker          B
3  Damon          O
4 Sophie         AB
> test2 <- data.frame(name = c('Damon','jimmy','nicker','tony'),
+                     group = c("group1","group1","group2","group2"),
+                     vision = c(4.2,4.3,4.9,4.5))
> test2
    name  group vision
1  Damon group1    4.2
2  jimmy group1    4.3
3 nicker group2    4.9
4   tony group2    4.5
> library(dplyr)
> inner_join(test1,test2,by="name")#取交集——与merge存在区别
    name blood_type  group vision
1  jimmy          A group1    4.3
2 nicker          B group2    4.9
3  Damon          O group1    4.2
> right_join(test1,test2,by="name")#右连接（保留右表存在信息）
    name blood_type  group vision
1  jimmy          A group1    4.3
2 nicker          B group2    4.9
3  Damon          O group1    4.2
4   tony       <NA> group2    4.5
> full_join(test1,test2,by="name")#全连接（取全集）
    name blood_type  group vision
1  jimmy          A group1    4.3
2 nicker          B group2    4.9
3  Damon          O group1    4.2
4 Sophie         AB   <NA>     NA
5   tony       <NA> group2    4.5
> semi_join(test1,test2,by="name")#半连接行（左边取子集，保留右表存在的信息）
    name blood_type
1  jimmy          A
2 nicker          B
3  Damon          O
> anti_join(test1,test2,by="name")#反连接
    name blood_type
1 Sophie         AB

五、字符串处理⭐️⭐️⭐️⭐️⭐️⭐️——stringr包

library(stringr)
x <- "The birch canoe slid on the smooth planks."
#检测字符串长度
str_length(x) #标点空格都算
[1] 42
#字符串拆分——str_split
x2 = str_split(x," ")[[1]];x2#以空格拆分，提取列表中元素（两个[])
x2 = str_split(x," ",simplify = T)[1,] #simplify将列表简化为矩阵
#字符串连接——str_c
str_c(x2,collapse = " ")#连接
str_c(x2,1234,sep = "+")#外部连接
#提取字符串的一部分
> str_sub(x,5,9)
[1] "birch"
#字符定位
> str_locate(x2,"n")
     start end
[1,]    NA  NA
[2,]    NA  NA
[3,]     3   3
[4,]    NA  NA
[5,]     2   2
[6,]    NA  NA
[7,]    NA  NA
[8,]     4   4
#字符检测!!!!!
str_detect(x2,"n")#检测是否含有“n”，并返回T/F
[1] FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE  TRUE
#与sum和mean连用，可以统计匹配的个数和比例
sum(str_detect(x2,"n"))
mean(str_detect(x2,"n"))
#字符串替换
str_replace(x2,"o","A")#默认替换每个字符串中匹配第一个
[1] "The"     "birch"   "canAe"   "slid"    "An"      "the"     "smAoth" 
[8] "planks."
str_replace_all(x2,"o","A")#替换全部
[1] "The"     "birch"   "canAe"   "slid"    "An"      "the"     "smAAth" 
[8] "planks."
#提取匹配到的字符
> str_extract(x2,"o|e")#默认提取每个字符串中匹配第一个，没有返回NA
[1] "e" NA  "o" NA  "o" "e" "o" NA 
> str_extract_all(x2,"o|e")
[[1]]
[1] "e"
[[2]]
character(0)
[[3]]
[1] "o" "e"
[[4]]
character(0)
[[5]]
[1] "o"
[[6]]
[1] "e"
[[7]]
[1] "o" "o"
[[8]]
character(0)
> str_extract_all(x2,"o|e",simplify = T)
     [,1] [,2]
[1,] "e"  ""  
[2,] ""   ""  
[3,] "o"  "e" 
[4,] ""   ""  
[5,] "o"  ""  
[6,] "e"  ""  
[7,] "o"  "o" 
[8,] ""   ""  
#字符删除
str_remove(x," ")#删除第一个空格字符
str_remove_all(x," ")

六、条件语句与循环语句⭐️⭐️⭐️⭐️⭐️⭐️——类似于C语言

1、if条件语句

（1）与else放一起使用

（2）ifelse

（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"))

2、for循环语句

（1）两种循环方式

（2）循环结果保存

s = 0
> x <- c(5,6,0,3)
> result = list()
> for(i in 1:length(x)){
+   s=s+x[[i]]
+   result[[i]] = c(x[[i]],s)
+ }
> result
[[1]]
[1] 5 5
[[2]]
[1]  6 11
[[3]]
[1]  0 11
[[4]]
[1]  3 14
> do.call(cbind,result)
     [,1] [,2] [,3] [,4]
[1,]    5    6    0    3
[2,]    5   11   11   14

七、矩阵、数据框的隐式循环——apply函数

八、列表的隐式循环——lapply()、sapply()

#lapply()返回值是列表
lapply(test,mean)
#sapply 简化结果，直接返回矩阵或向量
sapply(test,mean)