如何创建基于 N 个百分位的二进制向量？ 更新数据

如何解决如何创建基于 N 个百分位的二进制向量？ 更新数据

如何为高 25% 基因表达编码为 1 而所有其他患者编码为 0 的患者创建二元向量（命名为 group）？

> dput(head(dat,20))
c(14.0217647549219,4.38634192539018,11.230612647966,13.5882888840484,10.2699597878478,8.09562488203986,14.1224780341231,10.4488388145038,12.2745444001468,9.09203349810451,14.3513862469323,11.5782968747535,13.6411144041398,9.79892114560863,11.1019611651618,12.5146158084875,12.643970834391,1.09720624597437,5.83979838350692,11.1604484254692
)

解决方法

我们可以使用 quantile 创建群组

dat$group <- with(dat,+(V1 >= quantile(V1,0.75)))
dat$group
[1] 0 1 0 0 0 0 1 0

更新

根据评论，OP 的 'dat' 是 vector

dat <-  c(14.0217647549219,4.38634192539018,11.230612647966,13.5882888840484,10.2699597878478,8.09562488203986,14.1224780341231,10.4488388145038,12.2745444001468,9.09203349810451,14.3513862469323,11.5782968747535,13.6411144041398,9.79892114560863,11.1019611651618,12.5146158084875,12.643970834391,1.09720624597437,5.83979838350692,11.1604484254692 )

因此，我们可以直接在对象上应用quantile

group <- +(dat >= quantile(dat,0.75))
out <- data.frame(V1 = dat,group)

-输出

 out
          V1 group
1  14.021765     1
2   4.386342     0
3  11.230613     0
4  13.588289     1
5  10.269960     0
6   8.095625     0
7  14.122478     1
8  10.448839     0
9  12.274544     0
10  9.092033     0
11 14.351386     1
12 11.578297     0
13 13.641114     1
14  9.798921     0
15 11.101961     0
16 12.514616     0
17 12.643971     0
18  1.097206     0
19  5.839798     0
20 11.160448     0

数据

dat <- structure(list(V1 = c(4.124,5.215,1.368,0.325,0.368,3.653,36.12,0.124)),class = "data.frame",row.names = c("1","2","3","4","5","6","7","8"))