生信入门 第六天

options("repos"=c(CRAN="http://mirrors.tuna.tsinghua.edu.cn/CRAN/"))
options(BioC_mirror="http://mirrors.tuna.tsinghua.edu.cn/bioconductor/")

install.packages("dplyr)
library(dplyr)

test <- iris[c(1:2,51:52,101:102),]
View(test)

mutate(test, new = Sepal.Length * Sepal.Width) # add a new colunm named "new" ,add to the right of the table test, then print the whole table.

select(test, 1)  #pick the 1st colunm of table test, print on the screen.
select(test, c(1,5)) #pick the 1st and 5th column of test, print them on the screen.
select(test, Petal.Length, Petal.Width) # pick by column names.

vars <- c("Petal.Length", "Petal.Width") # define a vector named c, including those two colunm names
select(test, one_of(vars)) # one_of

filter(test, Species == "setosa") # in Species column, pick the rows which Species column equals "setosa",  == 是判断第一个向量的每个元素是否等于第二个向量的相对应元素，返回逻辑值
filter(test, Species == "setosa"&Sepal.Length > 5) # 选Species是setosa同时Sepal.Length 大于5的行, & 是 and
filter(test, Species %in% c("setosa","versicolor")) # 选Species 是 setosa和versicolor的行， %in% 是用于判断前一个向量的元素是否在后一个向量中，返回逻辑值。**这句用来查询并抽提数据会非常有用！**

arrange(test, Sepal.Length) #按照Sepal.Length的值，重排列行，默认从小到大排序
arrange(test, desc(Sepal.Length)) #用desc从大到小

summarise(test, mean(Sepal.Length), sd(Sepal.Length)) # mean()计算Sepal.Length的平均值, sd()计算标准差
# 一个组合应用实例： 先按照Species分组，再计算每组Sepal.Length的平均值和标准差
group_by(test, Species) 
summarise(group_by(test, Species),mean(Sepal.Length), sd(Sepal.Length))

test %>% 
  group_by(Species) %>% 
  summarise(mean(Sepal.Length), sd(Sepal.Length))

count(test, Species)

test1 <- data.frame(x = c('b','e','f','x'), 
                    z = c("A","B","C",'D'))
test1
test2 <- data.frame(x = c('a','b','c','d','e','f'), 
                    y = c(1,2,3,4,5,6))
test2 

inner_join(test1, test2, by = "x")  #內连取交集, 不能匹配的cases丢弃

left_join(test1, test2, by = 'x') # 左连，以前面左边这个test1的x为准，把test2的内容匹配过去，不匹配的cases丢弃
left_join(test2, test1, by = 'x') # 左连，以前边左边这个test2的x为准，把test的呢内容匹配进去，确实数值用NA，不匹配的丢弃

left_join(test2, test1, by = 'x') # 全连, 把test 1，test2 以x为准合并，长表变短表

semi_join(x = test1, y = test2, by = 'x') #半连接, 返回能够与y表匹配的x表所有记录，不合并两表格，只针对x操作

anti_join(x = test2, y = test1, by = 'x') # 反向半连接，返回不能够与y表匹配的x表所有记录，不合并两表格，只针对x操作

test1 <- data.frame(x = c(1,2,3,4), y = c(10,20,30,40))
test1
test2 <- data.frame(x = c(5,6), y = c(50,60))
test2
test3 <- data.frame(z = c(100,200,300,400))
test3
bind_rows(test1, test2)
bind_cols(test1, test3)
bind_rows(test1,test3) #列数不同，报错么？
bind_cols(test2,test3) #行数不同，如何？

?sd

browseVignettes("limma")