如何在R中基于多条件将一列值替换为另一列对应行的值

现有如下表格数据：

Family          Genus           Species                 Genus_new   col2
Pedinellales    Pedinella       uncultured_eukaryote    uncultured  eukaryote
Noctilucales    Noctiluca       Noctiluca_scintillans   Noctiluca   scintillans
Mediophyceae    Chaetoceros     Chaetoceros_socialis    Chaetoceros socialis
Noctilucales    Noctiluca       Noctiluca_scintillans   Noctiluca   scintillans
Eustigmatales   uncultured      uncultured_eukaryote    uncultured  eukaryote
Mediophyceae    Chaetoceros     Chaetoceros_sp.         Chaetoceros sp.
Gonyaulacales   Heterodinium    uncultured_eukaryote    uncultured  eukaryote
Cryptomonadales Cryptochloris   phytoplankton_environmental phytoplankton   environmental
Prymnesiophyceae    Phaeocystis Haptophyta_environmental    Haptophyta  environmental
Sarcinochrysidales  Pelagodinium   Pelagophyceae_sp.    Pelagophyceae   sp.
Mediophyceae    Skeletonema     uncultured_marine       uncultured  marine
Gymnodiniphycidae   Gyrodinium    uncultured_Gyrodinium   uncultured  Gyrodinium
Cryptomonadales Cryptochloris   uncultured_Cryptophyta  uncultured  Cryptophyta
Pedinellales    Pedinellales    uncultured_eukaryote    uncultured  eukaryote
Prasinophytae   Prasinophytae   Pycnococcus_sp.         Pycnococcus sp.

需求是：当Genus_new列的字符串满足以下任一条件时，将其替换为同一行Genus列的值：

• 以unc或phyto开头
• 以phyceae结尾

单独的for循环可以正常运行：

for (i in 1:nrow(data)) {
  (data$Genus_new[i][startsWith(data$Genus_new[i], "unc")]<-data$Genus[i])
} 
for (i in 1:nrow(data)) {
  (data$Genus_new[i][endsWith(data$Genus_new[i], "phyceae")]<-data$Genus[i])
}
for (i in 1:nrow(data)) {
  (data$Genus_new[i][startsWith(data$Genus_new[i], "phytoplank")]<-data$Genus[i])
}

但整合逻辑时遇到问题：

第一次尝试（行不匹配错误） #

#first try
data$Genus_new[startsWith(data$Genus_new, "unc") | 
                  startsWith(data$Genus_new[i], "Unc") | 
                  endsWith(data$Genus_new[i], "phyta") |
                  endsWith(data$Genus_new[i], "phyceae") |
                  startsWith(data$Genus_new[i], "phytoplank") |
                  endsWith(data$Genus_new[i], "phyte")
                ] <- data$Genus

结果出现行不匹配，比如第2行Genus_new的值变为第5行Genus的值。

第二次尝试（逻辑判断错误） #

#second try: 
for (i in 1:nrow(data)) {
  if(data$Genus_new[i][startsWith(data$Genus_new[i], "unc")]) {
    data$Genus_new[i]<-data$Genus[i]} 
   else if(data$Genus_new[i][endsWith(data$Genus_new[i], "phyta")]){
     data$Genus_new[i]<-data$Genus[i]} 
    else if(data$Genus_new[i][endsWith(data$Genus_new[i], "phyceae")]) {
      data$Genus_new[i]<-data$Genus[i]}
}

报错：

Error in if (data$Genus_new[i][startsWith(data$Genus_new[i], "unc")]) { : argument is not interpretable as logical

错误原因解析 #

1. 第一次尝试的问题：

   • 混用了向量化操作和循环变量i，startsWith(data$Genus_new[i], ...)是单个元素判断，而startsWith(data$Genus_new, "unc")是整列判断，导致生成的逻辑索引与data$Genus长度不匹配，赋值时出现行错位。
   • startsWith默认区分大小写，单独写"Unc"判断没必要，可用ignore.case = TRUE统一处理大小写。

2. 第二次尝试的问题：

   • data$Genus_new[i][startsWith(data$Genus_new[i], "unc")]写法错误：startsWith返回逻辑值时，该代码取的是data$Genus_new[i]的字符子集，而非逻辑值，if语句无法识别，因此报错。应直接用startsWith(data$Genus_new[i], "unc")作为判断条件。

正确实现方法 #

方法1：整合向量化操作（推荐，符合R语言风格） #

直接生成整列的逻辑索引，批量赋值避免循环：

# 生成所有满足条件的逻辑向量
condition <- startsWith(data$Genus_new, "unc", ignore.case = TRUE) |
             startsWith(data$Genus_new, "phytoplank") |
             endsWith(data$Genus_new, "phyceae") |
             endsWith(data$Genus_new, "phyta") |
             endsWith(data$Genus_new, "phyte")

# 对满足条件的行赋值
data$Genus_new[condition] <- data$Genus[condition]

方法2：整合for循环 #

修正第二次尝试的判断逻辑，直接用逻辑值作为判断条件：

for (i in 1:nrow(data)) {
  current_genus_new <- data$Genus_new[i]
  # 整合所有判断条件
  if (startsWith(current_genus_new, "unc") ||
      startsWith(current_genus_new, "phytoplank") ||
      endsWith(current_genus_new, "phyceae") ||
      endsWith(current_genus_new, "phyta") ||
      endsWith(current_genus_new, "phyte")) {
    data$Genus_new[i] <- data$Genus[i]
  }
}

方法3：正则表达式简化判断（更灵活） #

用grepl结合正则表达式一次性覆盖所有条件：

# 正则表达式：匹配以unc/phytoplank开头，或以phyceae/phyta/phyte结尾的字符串
condition <- grepl("^(unc|phytoplank)", data$Genus_new, ignore.case = TRUE) |
             grepl("(phyceae|phyta|phyte)$", data$Genus_new)

data$Genus_new[condition] <- data$Genus[condition]

内容的提问来源于stack exchange，提问作者Franchi