在tibble中从右列到左列移动数据

data <- tibble(
  id = c(1:10),
  diagnosis_1 = c("F32", "F431", "R58", "S32", "F11", NA, NA, "Y67", "F32", "Z032"),
  diagnosis_2 = c(NA, NA, NA, NA, NA, NA, "G35", NA, NA, NA),
  diagnosis_3 = c("F40", NA, "R67", "F431", NA, "F60", "S58", "R68", "F11", NA),
  diagnosis_4 = c(NA, NA, "F65", NA, "F19", NA, NA, "F32", NA, NA)
)

作为清洁过程的一部分,我已经删除了所有不符合某些标准的诊断(即不以字母F,G或Z开头).使用以下代码：

data$diagnosis_1[str_sub(data$diagnosis_1, 1,1) %in% c("R", "S", "Y")] <- NA
data$diagnosis_2[str_sub(data$diagnosis_2, 1,1) %in% c("R", "S", "Y")] <- NA
data$diagnosis_3[str_sub(data$diagnosis_3, 1,1) %in% c("R", "S", "Y")] <- NA
data$diagnosis_4[str_sub(data$diagnosis_4, 1,1) %in% c("R", "S", "Y")] <- NA

结束这个tibble：

我现在需要将数据向左移动以从左到右填充列(即,如果诊断_2,诊断_3或诊断_4具有数据,则诊断_1不为空).我已经尝试使用ifelse(),因为它是矢量化但我似乎无法使用几个嵌套的ifelse().

ifelse(is.na(data$diagnosis_1), data$diagnosis_2, data$diagnosis_1))

所有建议都非常感谢.

编辑：添加预期输出：

data[-1] <- lapply(data[-1], function(x) replace(x, grepl("^[R|S|Y]", x), NA))  
data[] <- t(apply(data, 1, function(x) `length<-`(na.omit(x), length(x))))

data
# A tibble: 10 x 5
#     id diagnosis_1 diagnosis_2 diagnosis_3 diagnosis_4
#   <chr> <chr>       <chr>       <chr>       <chr>      
# 1 " 1"  F32         F40         NA          NA         
# 2 " 2"  F431        NA          NA          NA         
# 3 " 3"  F65         NA          NA          NA         
# 4 " 4"  F431        NA          NA          NA         
# 5 " 5"  F11         F19         NA          NA         
# 6 " 6"  F60         NA          NA          NA         
# 7 " 7"  G35         NA          NA          NA         
# 8 " 8"  F32         NA          NA          NA         
# 9 " 9"  F32         F11         NA          NA         
#10  10   Z032        NA          NA          NA

将非NA值向左移动已经取自大卫从here开始的答案.你可以尝试任何其他方法来改变同一问题的值.