如何将R中以/分隔的基因ID替换为对应基因符号？

示例数据准备 #

先定义输入的基因ID向量和映射数据框：

geneIDs <- c("100/1000/100008586","1277/63923/8516","1133/1132/1956/8516")

gene_symbols <- data.frame(
  ENTREZID = c(1, 10, 100, 1000, 10000, 100008586, 1277, 63923, 8516, 1133, 1132, 1956),
  SYMBOL = c("A1BG", "NAT2", "ADA", "CDH2", "AKT3", "GAGE12F", "RAF1", "FOXP2", "CXCR4", "CSF1R", "CSF1", "JUN")
)

核心处理步骤 #

1. 统一ID格式：将映射表中的ENTREZID转为字符型，避免大数字因科学计数导致匹配失败：

gene_symbols$ENTREZID <- as.character(gene_symbols$ENTREZID)

2. 自定义转换函数：实现拆分ID、匹配符号、重新合并的逻辑：

convert_gene_ids <- function(id_str, mapping) {
  # 拆分斜杠分隔的ID
  ids <- strsplit(id_str, "/")[[1]]
  # 匹配对应的基因符号
  symbols <- mapping$SYMBOL[match(ids, mapping$ENTREZID)]
  # 可选：将未匹配的ID保留原值，避免出现NA
  symbols[is.na(symbols)] <- ids[is.na(symbols)]
  # 重新合并为斜杠分隔的字符串
  paste(symbols, collapse = "/")
}

3. 批量转换：将函数应用到整个基因ID向量：

geneSymbols <- sapply(geneIDs, convert_gene_ids, mapping = gene_symbols)

# 查看结果
geneSymbols

运行后输出：

#> "ADA/CDH2/GAGE12F" "RAF1/FOXP2/CXCR4" "CSF1R/CSF1/JUN/CXCR4"

Tidyverse风格实现（适合数据框列处理） #

如果是处理数据框中的列，用tidyverse工具链更便捷：

library(tidyverse)

df <- tibble(geneIDs = geneIDs)

df_processed <- df %>%
  mutate(
    geneSymbols = str_split(geneIDs, "/") %>%
      map(~ gene_symbols$SYMBOL[match(.x, gene_symbols$ENTREZID)]) %>%
      map_chr(paste, collapse = "/")
  )

print(df_processed)

输出结果：

#> # A tibble: 3 × 2
#>   geneIDs                  geneSymbols          
#>   <chr>                    <chr>                
#> 1 100/1000/100008586       ADA/CDH2/GAGE12F     
#> 2 1277/63923/8516          RAF1/FOXP2/CXCR4     
#> 3 1133/1132/1956/8516      CSF1R/CSF1/JUN/CXCR4

内容的提问来源于stack exchange，提问作者Victor Hugo Calegari de Toledo