在你做了连接之后，就用这个。它将检查空值并将该行的gene_id放入gene_name中：

new_df$gene_name <- ifelse(df$gene_name == '',df$gene_id,df$gene_name)

Tidyverse解决方案：我们可以在用NA替换空字符后使用coalesce，也可以使用replace。

library(dplyr)

left_join(df1, df2, by = "gene_id") %>%
    mutate(gene_name = na_if(gene_name, "")) %>%
    mutate(gene_name = coalesce(gene_name, gene_id)

#or

left_join(df1, df2, by = "gene_id") %>%
    mutate(gene_name = replace(gene_name,
                         gene_name == "",
                         gene_id[gene_name == ""]
                         )
           )

示例数据

df  <- tibble(id = paste0("some_string_",1:3), dat = c("bob", "mike", ""))

# A tibble: 3 × 2
  id            dat   
  <chr>         <chr> 
1 some_string_1 "bob" 
2 some_string_2 "mike"
3 some_string_3 ""

示例解决方案

df %>% 
    mutate(dat = na_if(dat, "")) %>% 
    mutate(dat = coalesce(dat, as.character(id)))

# A tibble: 3 × 2
  id            dat          
  <chr>         <chr>        
1 some_string_1 bob          
2 some_string_2 mike         
3 some_string_3 some_string_3