将多个数据帧粘贴在一起，同时将数据保留在特定的行和列中

Question

所以现在我有一个主数据框，其中包含230行和15132列（均标有特定的列和行名称），每个观察值中存储的值为0。我有多个具有不同列和行的数据框，我希望将它们粘贴到此主数据框中，同时保持正确的位置。这些数据帧中的每一个都由主数据帧的行/列名称的各种组合组成。我将如何去做呢？

dempty <- data.frame(matrix(0, nrow = 5, ncol = 5))
rownames(dempty) <- c("v", "w", "x", "y", "z")
colnames(dempty) <- c("a", "b", "c", "d", "e")


d1 <- data.frame(matrix(1, nrow = 2, ncol = 3))
rownames(d1) <- c("x", "y")
colnames(d1) <- c("b", "c", "d")

d2 <- data.frame(matrix(2, nrow = 3, ncol = 4))
rownames(d2) <- c("v", "w", "x")
colnames(d2) <- c("a", "b", "c", "d")

d3 <- data.frame(matrix(3, nrow = 4, ncol = 2))
rownames(d3) <- c("w", "x", "y", "z")
colnames(d3) <- c("d", "e")

dfinal <- data.frame(matrix(c(2, 2, 2, 0, 0, 2, 2, "1;2", "1;2", 0, 2, 2, "1;2", "1;2", 0, 2, "2;3", "1;2;3", "1;3", 3, 0, 3, 3, 3, 3), nrow = 5, ncol= 5))
rownames(dfinal) <- c("v", "w", "x", "y", "z")
colnames(dfinal) <- c("a", "b", "c", "d", "e")

Answer 1

此循环应使用dfinal，df1和df2中的值填充df3。可以根据您的实际问题轻松对其进行编辑。

## get row an col names of master data to be populated
cols<-names(dfinal)
rows<-row.names(dfinal)
## iterate through the master data frame
for(i in 1:length(cols)){
  for(j in 1:length(rows)){
    ## if df1 contains matching col/row names use ifelse to replace
    if(cols[i] %in% names(df1) & rows[j] %in% row.names(df1)){
      ## if current row/col in dfinal is still 0 replace with df1 value, else append df1 value
      dfinal[rows[j],cols[i]] <- ifelse(dfinal[rows[j],cols[i]] == 0, df1[rows[j],cols[i]], paste0(dfinal[rows[j],cols[i]],"; ",df1[rows[j],cols[i]]))
    }
    ## if current row/col in dfinal is still 0 replace with df2 value, else append df2 value
    if(cols[i] %in% names(df2) & rows[j] %in% row.names(df2)){
      ## if current row/col in dfinal is still 0 replace with df2 value, else append df2 value
      dfinal[rows[j],cols[i]] <- ifelse(dfinal[rows[j],cols[i]] == 0, df2[rows[j],cols[i]], paste0(dfinal[rows[j],cols[i]],"; ",df2[rows[j],cols[i]]))
    }
    ## if current row/col in dfinal is still 0 replace with df3 value, else append df3 value
    if(cols[i] %in% names(df3) & rows[j] %in% row.names(df3)){
      ## if current row/col in dfinal is still 0 replace with df3 value, else append df3 value
      dfinal[rows[j],cols[i]] <- ifelse(dfinal[rows[j],cols[i]] == 0, df3[rows[j],cols[i]], paste0(dfinal[rows[j],cols[i]],"; ",df3[rows[j],cols[i]]))
    }
  }
}

我希望这会有所帮助！

Answer 2

使用此示例数据：

dat <- `dimnames<-`(data.frame(matrix(0, nrow=3, ncol=5)), list(letters[1:3], LETTERS[1:5]))
dat
#   A B C D E
# a 0 0 0 0 0
# b 0 0 0 0 0
# c 0 0 0 0 0

df1 <- `rownames<-`(data.frame(B=2:3, C=22:23), c("a", "c"))
df1
#   B  C
# a 2 22
# c 3 23

您可以使用以下方法强行替换：

dat[,colnames(df1)] <- Map(function(x, y, rn) replace(x, rn, y),
                           dat[,colnames(df1)], df1,
                           list(match(rownames(df1), rownames(dat))))
dat
#   A B  C D E
# a 0 2 22 0 0
# b 0 0  0 0 0
# c 0 3 23 0 0

但是，如果您的数据具有作为实际列的行索引（行名），则对merge可能更容易且更可靠，或者使用dplyr::*_join函数之一进行某些合并后清除。基R的merge确实支持按行名合并，尽管感觉不太流畅：

newdat <- merge(dat, df1, by=0, all=TRUE)
newdat
#   Row.names A B.x C.x D E B.y C.y
# 1         a 0   2  22 0 0   2  22
# 2         b 0   0   0 0 0  NA  NA
# 3         c 0   3  23 0 0   3  23

(cleanupvars <- gsub("\\.x$", "", grep("\\.x$", colnames(newdat), value = TRUE)))
# [1] "B" "C"
for (nm in cleanupvars) {
  newdat[[nm]] <- ifelse(is.na(newdat[[paste0(nm, ".y")]]), newdat[[paste0(nm, ".x")]], newdat[[paste0(nm, ".y")]])
  newdat[,paste0(nm, c(".x", ".y"))] <- NULL
}
newdat
#   Row.names A D E B  C
# 1         a 0 0 0 2 22
# 2         b 0 0 0 0  0
# 3         c 0 0 0 3 23
newdat$Row.names <- NULL

（不保留列顺序。）

毫无疑问，这最后一种方法很笨重。