＃數據1如何使用「merge，by = Column.name」函數合併具有不同名稱的列？
SampleID <- c("A-01","B-01","C-01")
Value <- c(1,2,3)
data1 <- data.frame(SampleID, Value)
＃數據2
SampleID <- c("A","B","C")
Value1 <- c(3,4,5)
data2 <- data.frame(SampleID,Value1)
＃輸出：我想是使用以下： merge(data1, data2, by=c("SampleID"), all = TRUE)
SampleID Value Value1
A-01 1 3
B-01 2 4
C-01 3 5
刪除data1多餘的欄，可以首先從數據1分裂SampleID，然後拼接它。
SampleID <- c("A-01","B-01","C-01")
Sample <- substr(SampleID,1,1)
Num <- substr(SampleID,3,5)
Value <- c(1,2,3)
data1 <- data.frame(Sample ,Num, Value)
SampleID <- c("A","B","C")
Value1 <- c(3,4,5)
data2 <- data.frame(SampleID, Value1)
merged <- merge(data1, data2, by.x = "Sample", by.y = "SampleID", all = T)
merged$SampleID <- paste(merged$Sample,merged$Num, sep = "-")
merged <- merged[,c(5,3,4)]
SampleID Value Value1
1 A-01 1 3
2 B-01 2 4
3 C-01 3 5
可以使用sqldf庫做到這一點：
library(sqldf);
sqldf("SELECT data1.SampledId, data1.Vlaue, data2.Value2 FROM data1 JOIN data2 on data1.SampleID like data1.SampleID + '-%'")
或者使用data.table喜歡以下內容：
library(data.table)
dt1 <- data.table(data1)
dt2 <- data.table(data2)
dt1[dt2, on = .(grepl(CustomerId, CustomerId)), all = TRUE]
我相信下面做你所需要的。
data1$NewID <- gsub("[^[:alpha:]]", "", data1$SampleID)
result <- merge(data1, data2, by.x = "NewID", by.y = "SampleID", all = TRUE)
result <- result[-1]
result
# SampleID Value Value1
#1 A-01 1 3
#2 B-01 2 4
#3 C-01 3 5
可以再用
data1 <- data1[-3]
要添加到收藏，這裏是一個dplyr解決方案，讀取更容易一點：
options(stringsAsFactors = F)
SampleID <-c("A-01","B-01","C-01")
Value <- c(1,2,3)
data1 <- data.frame(SampleID, Value)
SampleID <- c("A","B","C")
Value1 <- c(3,4,5)
data2 <- data.frame(SampleID,Value1)
data1 %>%
mutate(new_id = gsub("[^[:alpha:]]", "", SampleID)) %>%
left_join(., data2, by = c("new_id" = "SampleID")) %>%
select(-new_id)
SampleID Value Value1
1 A-01 1 3
2 B-01 2 4
3 C-01 3 5
使用參數'by.x'和'by.y'。 –
所以你想合併行'A-01'和行'A'等？如果是這樣，你必須首先使這些值相等，用'gsub'創建一個新列。 –