在 R 函数中解析变量的问题

Question

`我正在使用 R 中的数据集，我想创建一个函数，根据四分位数将值分为三个类别。低于第一四分位数 (Q1) 的值应标记为“低表达”，高于第三四分位数 (Q3) 的值应标记为“高表达”，其他所有值应标记为“NA”。

我编写了以下函数，但在函数内部，过滤步骤似乎忽略了具有低值（低于 Q1）和 NA 值的行。有人可以帮我找出问题所在吗？

分离四分位数的函数

fun_1 <- function(df, var_1) {
  
    # separating quartiles
    quant_25 <- quantile(df[[var_1]], 0.25, names = F)
    quant_75 <- quantile(df[[var_1]], 0.75, names = F)
  

    # selecting columns of importance and creating a new column classifying quartiles
    tmp_1 <- df %>%
      select(c(1:7) | {{var_1}}) %>%
      mutate(
        expression =
          case_when(
            {{var_1}} <= quant_25 ~ paste(
              "Low expression ( < ", quant_25, ")"),
            {{var_1}} >= quant_75 ~ paste(
              "High expression ( >= ", quant_75, ")"),
            TRUE ~ "NA")) %>%
      filter(expression != "NA")
  
    tmp_1
}

通话功能

fun_1(dataset_name, "gene_name")

运行代码后，我得到以下输出：函数的输出：

但是，我希望该函数删除第一个 (Q1) 和第三个四分位数 (Q3) 之间的值，包括低表达值（低于 Q1）。

编辑：

数据集示例：

colnames <- c("sample_id", "cell_type", "treatment", "replicate", "time_point", "read_count", "normalized_count", "fpkm", "gene_name")

dataset_name <- data.frame(matrix(nrow = 20, ncol = length(colnames)))

for (i in 1:length(colnames)) {
  dataset_name[,i] <- ifelse(i == length(colnames), 
                              round(runif(20, min = 0.1, max = 100), digits = 3),
                              round(runif(20, min = 1, max = 100), digits = 3))
}

dataset_name <- dataset_name[, c(1:(length(colnames)-1), length(colnames))]

print(dataset_name)

Answer 1

fun_2 <- function(df, var_1) {
  library(dplyr)
  quants <- quantile(df[[var_1]], c(0.25, 0.75), names = FALSE)
  df %>% select(1:7 | {{var_1}}) %>%
    mutate(expression = case_when(
      {{var_1}} < quants[1] ~ sprintf("Low expression ( < %.02f )", quants[1]),
      {{var_1}} >= quants[2] ~ sprintf("High expression ( >= %.02f )", quants[2]),
      .default = "NA")
    ) %>%
    filter(expressions != "NA")
}
fun_2(dataset_name, "X9")
#        X1     X2    X3     X4     X5     X6     X7     X9                   expression
# 1  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 2  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 3  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 4  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 5  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 6  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 7  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 8  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 9  75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 10 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 11 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 12 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 13 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 14 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 15 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 16 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 17 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 18 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 19 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )
# 20 75.782 58.059 8.712 45.323 76.409 83.199 84.955 84.366 High expression ( >= 84.37 )

意识到

filter

没有执行任何操作，因为没有任何数据（在此数据集中的任何列中）位于四分位数之间。

---

您的样本数据：

dataset_name <- structure(list(X1 = c(75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782, 75.782), X2 = c(58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059, 58.059), X3 = c(8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712, 8.712), X4 = c(45.323,  45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323, 45.323), X5 = c(76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409, 76.409), X6 = c(83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199, 83.199), X7 = c(84.955,  84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955, 84.955), X8 = c(86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255, 86.255), X9 = c(84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366, 84.366)), row.names = c(NA,  -20L), class = "data.frame")