在R中用Loop/apply批量计算配对列差值生成新列

问题说明 #

现有包含多组_OVX/_SAC配对测量列的数据框，需要批量为每组变量生成新列，计算对应_SAC列减去_OVX列的差值。示例数据如下：

df <- structure(list(Leptin_OVX = c(101537.030773452, 34184.1969018313, 
54567.1867690491, 29558.5420636246, 40929.680418857), Leptin_SAC = c(19785.945743781, 
124224.32770312, 89539.7367479193, 29335.4677793977, 49085.7085270132
), MIP1A_OVX = c(198.955714001384, 9.39084362457698, 6.31036668689314, 
31.4248133610863, 61.7242016227428), MIP1A_SAC = c(152.595958885867, 
0, 6.31036668689314, 12.0518867341972, 56.3458462409656), IL4_OVX = c(84.3973038052031, 
0, 0, 84.3973038052031, 0), IL4_SAC = c(0, 0, 0, 0, 0), IL1B_OVX = c(20.5433459761151, 
0, 0, 0, 26.9522602664794), IL1B_SAC = c(0, 18.9503177384518, 
14.986896887192, 0, 0)), row.names = c(NA, -5L), class = c("tbl_df", 
"tbl", "data.frame"))

目前可通过手动多次调用mutate实现，但需要更高效的批量处理方案：

df <- df %>%
  mutate(chg_Leptin = Leptin_SAC - Leptin_OVX) %>%
  mutate(chg_MIP1A = MIP1A_SAC - MIP1A_OVX) %>%
  mutate(chg_IL4 = IL4_SAC - IL4_OVX) %>%
  mutate(chg_IL1B = IL1B_SAC - IL1B_OVX)

解决方案 #

方法1：使用tidyverse（dplyr）批量处理（推荐） #

利用dplyr::across实现向量式批量操作，无需循环：

library(dplyr)

# 提取所有配对变量的前缀（去除_OVX/_SAC后缀）
var_prefixes <- unique(sub("_OVX|_SAC", "", colnames(df)))

# 批量生成差值列
df <- df %>%
  mutate(
    across(
      .cols = all_of(paste0(var_prefixes, "_SAC")),
      .fns = ~ .x - get(sub("_SAC", "_OVX", cur_column())),
      .names = "chg_{sub('_SAC', '', .col)}"
    )
  )

方法2：Base R循环实现 #

如果偏好使用循环，可通过遍历变量前缀完成批量计算：

# 提取变量前缀
var_prefixes <- unique(sub("_OVX|_SAC", "", colnames(df)))

# 循环生成差值列
for (prefix in var_prefixes) {
  sac_col <- paste0(prefix, "_SAC")
  ovx_col <- paste0(prefix, "_OVX")
  df[[paste0("chg_", prefix)]] <- df[[sac_col]] - df[[ovx_col]]
}

内容的提问来源于stack exchange，提问作者Erin Giles