r - 根据 id 序列扩展列

Question

我想从包含 ID 序列的列创建新列，比如 ID1-5。

比如说，我们有以下小标题:

# A tibble: 3 x 2
  group   id_seq   
  <chr>   <chr>       
1 A       ID_Nr61-63
2 A       ID_Nr67-69
3 B       ID_Nr73-75

我想要的输出是:

# A tibble: 3 x 6
  group   id_seq      id1      id2      id3      id4
  <chr>   <chr>       <chr>    <chr>    <chr>    <chr>       
1 A       ID_Nr61-64  ID_Nr61  ID_Nr62  ID_Nr63  ID_Nr64    
2 A       ID_Nr67-69  ID_Nr67  ID_Nr68  ID_Nr69  NA
3 B       ID_Nr73-75  ID_Nr73  ID_Nr74  ID_Nr75  NA

感谢您的帮助!谢谢。

Answer 1

我们可以从 id_seq 中提取数字，通过 rowwise，得到序列 (:) 来扩展数据并从 'long' 整形为 'wide' pivot_wider

library(dplyr)
library(tidyr)
library(stringr)
library(data.table)
df1 %>% 
   mutate(s1 = as.numeric(str_extract(id_seq, "\\d+")), 
    s2 = as.numeric(str_extract(id_seq, "\\d+$"))) %>%
   mutate(rn = row_number()) %>% 
  rowwise %>% 
  summarise(rn, group, id_seq, se = (s1:s2)) %>% 
  mutate(id_seq2 = str_replace(id_seq, "\\d+-\\d+$", as.character(se)), 
   rn2 = str_c("id", rowid(rn)), se = NULL) %>%
  pivot_wider(names_from = rn2, values_from = id_seq2) %>%
  select(-rn)

-输出

# A tibble: 3 × 6
  group id_seq     id1     id2     id3     id4    
  <chr> <chr>      <chr>   <chr>   <chr>   <chr>  
1 A     ID_Nr61-64 ID_Nr61 ID_Nr62 ID_Nr63 ID_Nr64
2 A     ID_Nr67-69 ID_Nr67 ID_Nr68 ID_Nr69 <NA>   
3 B     ID_Nr73-75 ID_Nr73 ID_Nr74 ID_Nr75 <NA>   

---

或者使用base R

lst1 <- lapply(sub("-", ":", sub("ID_Nr", "", df1$id_seq)), 
    function(x) paste0("id_seq", eval(parse(text = x))))
 mx <- max(lengths(lst1))
m1 <- do.call(rbind, lapply(lst1, `length<-`, mx))
df1[paste0("id", seq_len(ncol(m1)))] <- m1
df1
  group     id_seq      id1      id2      id3      id4
1     A ID_Nr61-64 id_seq61 id_seq62 id_seq63 id_seq64
2     A ID_Nr67-69 id_seq67 id_seq68 id_seq69     <NA>
3     B ID_Nr73-75 id_seq73 id_seq74 id_seq75     <NA>

数据

df1 <- structure(list(group = c("A", "A", "B"), id_seq = c("ID_Nr61-64", 
"ID_Nr67-69", "ID_Nr73-75")), class = "data.frame", row.names = c("1", 
"2", "3"))