在R中按group分组统计每组内的unique gene数量

需求：根据提供的example数据集，统计每个group下的唯一gene数量，预期输出为：

• "Amino acid metabolism" 和 "Carbohydrate metabolism" 组的total值为2
• "Translation" 和 "Membrane transport" 组的total值为1

用户尝试的两段代码均未得到预期结果：

尝试代码1 #

library(tidyverse)
example %>% 
  group_by(group,gene) %>% 
  mutate(total = n()) %>%
  unique() -> count.viz

尝试代码2 #

example %>% 
  select(gene) %>%
  group_by(group) %>% 
  mutate(total = n()) %>%
  unique() -> count.viz

数据集的dput结构：

> dput(example)
structure(list(KO = c("K00639", "K12308", "K01881", "K16785", 
"K03639", "K00174"), Project = c("Ga0598246", "Ga0598246", "Ga0598246", 
"Ga0598246", "Ga0598246", "Ga0598246"), count = c(0, 0, 0, 0, 
0, 0.0170424288146159), gene = c("kbl, GCAT", "bgaB, lacA", "PARS, proS", 
"ecfT", "moaA, CNX2", "korA, oorA, oforA"), group = c("Amino acid metabolism", 
"Carbohydrate metabolism", "Translation", "Membrane transport", 
"Amino acid metabolism", "Carbohydrate metabolism")), class = c("grouped_df", 
"tbl_df", "tbl", "data.frame"), row.names = c(NA, -6L), groups = structure(list(
    Project = "Ga0598246", .rows = structure(list(1:6), ptype = integer(0), class = c("vctrs_list_of", 
    "vctrs_vctr", "list"))), class = c("tbl_df", "tbl", "data.frame"
), row.names = c(NA, -1L), .drop = TRUE))  

错误代码分析 #

• 代码1：按group和gene联合分组后，n()统计的是每个(group, gene)组合的出现次数，而非每个group下的唯一gene总数，最终结果会保留所有gene行，不符合需求。
• 代码2：先执行select(gene)丢失了group列，后续group_by(group)会直接报错；即使保留group列，n()统计的是该group下的总行数，而非唯一gene的数量。

正确实现方法 #

以下两种tidyverse方法均可得到预期结果：

方法1：先去重再分组计数 #

先保留group和gene的唯一组合，再按group统计数量：

library(tidyverse)
count.viz <- example %>%
  distinct(group, gene) %>% # 保留唯一的group-gene组合
  group_by(group) %>%
  summarize(total = n())

方法2：直接使用n_distinct()统计唯一值 #

在分组后，用n_distinct(gene)直接计算每个group下的唯一gene数量：

count.viz <- example %>%
  group_by(group) %>%
  summarize(total = n_distinct(gene))

两种方法的输出结果一致：

# A tibble: 4 × 2
  group                    total
  <chr>                    <int>
1 Amino acid metabolism        2
2 Carbohydrate metabolism      2
3 Membrane transport           1
4 Translation                  1

内容的提问来源于stack exchange，提问作者Geomicro