平均 R 中的不同列

Question

我有一个像

> dput(head(b1))
structure(list(hgnc_symbol = c("DDX11L1", "WASH7P", "MIR6859-1", 
"MIR1302-2HG", "MIR1302-2", "FAM138A"), CS001_t1 = c(0L, 70L, 
9L, 0L, 0L, 0L), CS001_t2 = c(0L, 100L, 1L, 0L, 0L, 0L), CS001_1 = c(0L, 
80L, 1L, 0L, 0L, 0L), CS001_2 = c(5L, 86L, 2L, 0L, 0L, 0L), CS002_1 = c(1L, 
82L, 14L, 1L, 0L, 0L), CS002_2 = c(2L, 78L, 15L, 1L, 0L, 0L), 
    HC002_1 = c(4L, 83L, 21L, 0L, 0L, 0L), HC002_2 = c(8L, 94L, 
    14L, 0L, 0L, 0L), HC003_1 = c(17L, 101L, 18L, 0L, 0L, 0L), 
    HC003_2 = c(18L, 84L, 22L, 1L, 0L, 0L)), row.names = c("1", 
"2", "3", "4", "5", "6"), class = "data.frame")
> 

> names(b1)
 [1] "hgnc_symbol" "CS001_t1"    "CS001_t2"    "CS001_1"     "CS001_2"    
 [6] "CS002_1"     "CS002_2"     "HC002_1"     "HC002_2"     "HC003_1"    
[11] "HC003_2"    
> 

我想平均"CS001_t1" and "CS001_t2" 、 "CS001_1" and "CS001_2"等中的数字（四舍五入到十进制）

所以，最后我有five列

Answer 1

您可以使用split.default ：

cbind(b1[1], sapply(split.default(b1[-1], 
      sub('(.*_.*?)\\d$', '\\1', names(b1)[-1])), rowMeans, na.rm = TRUE))

#  hgnc_symbol CS001_ CS001_t CS002_ HC002_ HC003_
#1     DDX11L1    2.5       0    1.5    6.0   17.5
#2      WASH7P   83.0      85   80.0   88.5   92.5
#3   MIR6859-1    1.5       5   14.5   17.5   20.0
#4 MIR1302-2HG    0.0       0    1.0    0.0    0.5
#5   MIR1302-2    0.0       0    0.0    0.0    0.0
#6     FAM138A    0.0       0    0.0    0.0    0.0

sub部分用于区分每组列。

sub('(.*_.*?)\\d$', '\\1', names(b1)[-1])
#[1] "CS001_t" "CS001_t" "CS001_"  "CS001_"  "CS002_" 
#[6] "CS002_"  "HC002_"  "HC002_"  "HC003_"  "HC003_"