折疊每個家庭的資料

我有這個資料集，其中包含家庭中雙胞胎的值：

zyg   fid    x_t1    x_t2     y_t1   y_t2
 1 499474     NA     1      1    NA
 1 499474     NA     NA    NA    NA
 1 499474     NA     NA    NA     1
 1 499474     NA     NA    NA    NA
 1 499540     NA     NA     1    NA
 1 499540     NA     NA    NA    NA
 2 499874     NA     NA    NA    NA
 2 499874     NA     NA     1    NA
 2 499874     NA     NA    NA     1
 2 499874     2      NA    NA     1 

• 當存在 x 和 y 的表型資訊時，如何折疊保留這些表型資訊的家族？

家庭 499479 的預期是：

zyg   fid    x_t1    x_t2  y_t1   y_t2
 1 499474     NA     1      1     1

對于家庭 499874，它應該是：

 2 499874     2      NA    1     1 

uj5u.com熱心網友回復：

您可以使用以下代碼：

library(dplyr)

df %>%
  group_by(fid) %>%
  summarise_all(~first(na.omit(.)))

輸出：

# A tibble: 3 × 6
     fid   zyg  x_t1  x_t2  y_t1  y_t2
   <int> <int> <int> <int> <int> <int>
1 499474     1    NA     1     1     1
2 499540     1    NA    NA     1    NA
3 499874     2     2    NA     1     1

您的資料：

df<-structure(list(zyg = c(1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L
), fid = c(499474L, 499474L, 499474L, 499474L, 499540L, 499540L, 
499874L, 499874L, 499874L, 499874L), x_t1 = c(NA, NA, NA, NA, 
NA, NA, NA, NA, NA, 2L), x_t2 = c(1L, NA, NA, NA, NA, NA, NA, 
NA, NA, NA), y_t1 = c(1L, NA, NA, NA, 1L, NA, NA, 1L, NA, NA), 
    y_t2 = c(NA, NA, 1L, NA, NA, NA, NA, NA, 1L, 1L)), class = "data.frame", row.names = c(NA, 
-10L))

uj5u.com熱心網友回復：

如果只有一個非 NA 元素，則每組

library(dplyr)
df1 %>%
 group_by(zyg, fid) %>%
  summarise(across(everything(), ~ .x[complete.cases(.x)][1]), .groups = "drop")

-輸出

# A tibble: 3 × 6
    zyg    fid  x_t1  x_t2  y_t1  y_t2
  <int>  <int> <int> <int> <int> <int>
1     1 499474    NA     1     1     1
2     1 499540    NA    NA     1    NA
3     2 499874     2    NA     1     1

資料

df1 <- structure(list(zyg = c(1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 2L
), fid = c(499474L, 499474L, 499474L, 499474L, 499540L, 499540L, 
499874L, 499874L, 499874L, 499874L), x_t1 = c(NA, NA, NA, NA, 
NA, NA, NA, NA, NA, 2L), x_t2 = c(1L, NA, NA, NA, NA, NA, NA, 
NA, NA, NA), y_t1 = c(1L, NA, NA, NA, 1L, NA, NA, 1L, NA, NA), 
    y_t2 = c(NA, NA, 1L, NA, NA, NA, NA, NA, 1L, 1L)), 
class = "data.frame", row.names = c(NA, 
-10L))

uj5u.com熱心網友回復：

另一種可能的解決方案：

library(dplyr)

df %>% 
  group_by(zyg, fid) %>% 
  summarise(across(everything(), ~ if (all(is.na(.x))) {NA} else
           {max(.x, na.rm = T)}), .groups = "drop")

#> # A tibble: 3 × 6
#>     zyg    fid  x_t1  x_t2  y_t1  y_t2
#>   <int>  <int> <int> <int> <int> <int>
#> 1     1 499474    NA     1     1     1
#> 2     1 499540    NA    NA     1    NA
#> 3     2 499874     2    NA     1     1

uj5u.com熱心網友回復：

與其他答案非常相似，但也想給出我自己的解決方案。

df %>%
  group_by(zyg,fid) %>% 
  summarise(across(everything(),~sum(.,na.rm=TRUE))
            )

uj5u.com熱心網友回復：

您想做一些coalesce按行對列所做的事情：

您可以這樣做：

libarary(dplyr)

coalesce_by_column <- function(df) {
  return(dplyr::coalesce(!!! as.list(df)))
}

df %>%
  group_by(fid) %>%
  summarise(across(everything(), coalesce_by_column))

     fid   zyg  x_t1  x_t2  y_t1  y_t2
   <int> <int> <int> <int> <int> <int>
1 499474     1    NA     1     1     1
2 499540     1    NA    NA     1    NA
3 499874     2     2    NA     1     1

標籤：r dplyr

上一篇：rmarkdown中的編號gt表

下一篇：R：使用dplyr包獲取小標題的行號