如何收集用tidyr按顺序编号的列答案

【问题标题】：how to gather columns which are numbered sequentially with tidyr如何收集用tidyr按顺序编号的列
【发布时间】：2016-11-28 19:30:35
【问题描述】：

我有一组列如下

rowid   TimePoint2   TimePoint3    TimePoint4     TimePoint5    TimePoint6

我想整理数据，但是当我使用他的粘贴功能时，收集功能无法识别列名我想要的是

rowid
1   TimePoint2    43
2   TimePoint3    34
3   TimePoint4    24
4   TimePoint5    22
5   TimePoint6    44

等

我的代码：

所以不要写：

y="TimePoint"
  Mydf<-
    select(df,matches(y),rowid)%>%
    gather(variable, value, TimePoint1,TimePoint2,TimePoint3,TimePoint4,TimePoint5,TimePoint6)

我想写这样的东西：

y="TimePoint"
  Mydf<-
    select(df,matches(y),rowid)%>%
    gather(variable, value, paste(y,1:10,",",sep=""))

我得到的错误是Error: All select() inputs must resp;ve to integer column positions. The following do not: * paste (y,1:10,",",sep="")

【问题讨论】：

我认为您可以使用 dplyr 中的 starts_with，就像在 select() 中一样。
好的，很好。请添加为答案

标签： r tidyr

【解决方案1】：

我相信你可以做到：

Mydf <- select(df,matches(y),rowid)%>%
    gather(variable, value, starts_with("TimePoint"))

【讨论】：

【解决方案2】：

由于您正在选择与“时间点”匹配的列，这也将起作用：

library(dplyr)
library(tidyr)
y <- "TimePoint"
Mydf <- df %>% select(rowid,matches(y)) %>% 
               gather(variable,value,-1)
print(Mydf)
##  rowid   variable value
##1     1 TimePoint2    43
##2     1 TimePoint3    34
##3     1 TimePoint4    24
##4     1 TimePoint5    22
##5     1 TimePoint6    44

在这里，我们像您一样 select，但将 rowid 列放在首位。然后使用-1 收集除第一列之外的所有列。

数据：

df <- structure(list(rowid = 1L, TimePoint2 = 43L, TimePoint3 = 34L, 
    TimePoint4 = 24L, TimePoint5 = 22L, TimePoint6 = 44L), .Names = c("rowid", 
"TimePoint2", "TimePoint3", "TimePoint4", "TimePoint5", "TimePoint6"
), class = "data.frame", row.names = c(NA, -1L))
##  rowid TimePoint2 TimePoint3 TimePoint4 TimePoint5 TimePoint6
##1     1         43         34         24         22         44

【讨论】：