r语言 - 如何将列中的多个分隔字符串分割为一行，并删除该行id中的重复项?

Question

我的数据帧是这样的

tbody> <<tr>

Fasta标头
ab12_P002; ab12_P003; ab12_P005; ab23_P002; ab23_P001
ab45_P001; ab36_P001
ab55_P001; ab55_P002

这里有一个稍微不同的方法:

library(stringr)
library(dplyr)
library(tidyr)
without_02473 %>% 
separate_rows(`Fasta headers`) %>% 
filter(str_detect(`Fasta headers`, 'ab\d+')) %>% 
distinct()

`Fasta headers`
<chr>          
1 ab12           
2 ab23           
3 ab45           
4 ab36           
5 ab55  

使用strsplit并删除_之后的所有内容并过滤distinct的另一个选项如下:

library(dplyr)
library(tidyr)
without_02473 %>%
mutate(`Fasta headers` = strsplit(`Fasta headers`, ";")) %>%
unnest(`Fasta headers`) %>%
mutate(`Fasta headers` = sub("_[^_]+$", "", `Fasta headers`)) %>%
distinct()
#> # A tibble: 5 × 1
#>   `Fasta headers`
#>   <chr>          
#> 1 ab12           
#> 2 ab23           
#> 3 ab45           
#> 4 ab36           
#> 5 ab55

^{创建于2023-01-03与reprex v2.0.2}

我们可以使用separate_rows在;处拆分Fasta headers以创建新行，然后使用trimws从_开始删除后缀部分

library(dplyr)
library(tidyr)
out <- without_02473 %>% 
separate_rows(`Fasta headers`, sep = ";") %>%
mutate(`Fasta headers` = trimws(`Fasta headers`, whitespace = "_.*")) %>%
distinct

与产出

out
# A tibble: 5 × 1
`Fasta headers`
<chr>          
1 ab12           
2 ab23           
3 ab45           
4 ab36           
5 ab55   
library(writexl) 
write_xlsx(out, "first.xlsx")

---

或者可以只提取_之前的单词，str_extract_all,unnestlist列，并获得distinct行

library(stringr)
without_02473 %>%
mutate(`Fasta headers` = str_extract_all(`Fasta headers`, 
"\w+(?=_)")) %>%
unnest(`Fasta headers`) %>%
distinct

数据

without_02473 <- structure(list(`Fasta headers` = c("ab12_P002;ab12_P003;ab12_P005;ab23_P002;ab23_P001", 
"ab45_P001;ab36_P001", "ab55_P001;ab55_P002")), class = "data.frame", row.names = c(NA, 
-3L))