由于要丢弃前面的S[0-9]-部分，因此我们使用names_pattern=并丢弃前面的部分。

pivot_longer(df, cols  = -c(gene_id, gene_short_name),
    names_to = ".value", names_pattern = ".*-(.*)")
# # A tibble: 3 × 4
#   gene_id gene_short_name  Case Control
#   <chr>   <chr>           <dbl>   <dbl>
# 1 EN0001  TFG             0.003   0.001
# 2 EN0001  TFG             0.001   0.002
# 3 EN0001  TFG             0.002   0.003

字符串

• 编辑 *：为了保留S[1-6]，这在这里可以工作，并且可能足以在一般情况下工作。我保留了辅助变量rn1和rn2主要是为了显示它们包含的内容，它们可以在第二个pivot之后安全地删除：

df %>%
  mutate(rn1 = row_number()) %>%
  pivot_longer(cols  = -c(rn1, gene_id, gene_short_name)) %>%
  separate(name, into = c("S", "var")) %>%
  group_by(rn1, var) %>%
  mutate(rn2 = row_number()) %>%
  ungroup() %>%
  pivot_wider(c(gene_id, gene_short_name, rn1, rn2), names_from = var, values_from = c(value, S))
# # A tibble: 3 × 8
#   gene_id gene_short_name   rn1   rn2 value_Case value_Control S_Case S_Control
#   <chr>   <chr>           <int> <int>      <dbl>         <dbl> <chr>  <chr>    
# 1 EN0001  TFG                 1     1      0.003         0.001 S1     S4       
# 2 EN0001  TFG                 1     2      0.001         0.002 S2     S5       
# 3 EN0001  TFG                 1     3      0.002         0.003 S3     S6

型
数据类型

df <- structure(list(gene_id = "EN0001", gene_short_name = "TFG", "S1-Case" = 0.003, "S2-Case" = 0.001, "S3-Case" = 0.002, "S4-Control" = 0.001, "S5-Control" = 0.002, "S6-Control" = 0.003), class = "data.frame", row.names = c(NA, -1L))

型

不是有效的代码，而是一种替代方法

df %>% pivot_longer(cols = c(starts_with('S'))) %>% 
  separate(name, into = c('s','c'), sep = '-') %>% 
  pivot_wider(id_cols = c(gene_id,gene_short_name), names_from = c, values_from = value, values_fn = list) %>% 
  unnest(c(Case, Control))

字符串
创建于2023-07-15带有reprex v2.0.2

# A tibble: 3 × 4
  gene_id gene_short_name  Case Control
  <chr>   <chr>           <dbl>   <dbl>
1 EN0001  TFG             0.003   0.001
2 EN0001  TFG             0.001   0.002
3 EN0001  TFG             0.002   0.003

型