Question

我有一个包含样本分类的数据框：

 Seq_ID   Family Father   Mother   Sex    Role    Type  
   <chr>     <dbl> <chr>    <chr>    <chr>  <chr>   <chr> 
 1 SSC02219 11000. 0        0        Male   Father  Parent
 2 SSC02217 11000. 0        0        Female Mother  Parent
 3 SSC02254 11000. SSC02219 SSC02217 Male   Proband Child 
 4 SSC02220 11000. SSC02219 SSC02217 Female Sibling Child 
 5 SSC02184 11001. 0        0        Male   Father  Parent
 6 SSC02181 11001. 0        0        Female Mother  Parent
 7 SSC02178 11001. SSC02184 SSC02181 Male   Proband Child 
 8 SSC03092 11002. 0        0        Male   Father  Parent
 9 SSC03078 11002. 0        0        Female Mother  Parent
10 SSC03070 11002. SSC03092 SSC03078 Female Proband Child

目前，从a到b，我必须这样做：

library(tidyverse)
library(janitor)

sample.df %>% tabyl(Role, Sex) %>% 
  adorn_totals(where=c("row", "col") ) %>% 
  as.tibble() %>% select(1,4,3,2) %>%
  # Part 2
  mutate(type=c("parent", "parent", "child", "child", " ")) %>% 
  inner_join(., group_by(., type) %>% 
  summarise(total=sum(Total))) %>% 
  select(5,6,1,2,3,4)

我觉得这是一个非常简单的解决方法。在dplyr中有更直接的方法来完成第二部分吗？

一

b

Answer 1

这是一个选项。 as.tibble没有必要。当你有很多课程分配给＆＃34; parent＆＃34;时，mutate case_when更易于管理或者＆＃34;孩子＆＃34;。我们不需要inner_join，因为我们可以使用group_by和mutate来计算total。最后，我喜欢在使用select函数时写下列名，因为将来我更容易阅读，但是你可以使用列索引，只要你确信列索引无论您在管道操作中包含哪些新分析，都将保持不变。

library(tidyverse)
library(janitor)

sample.df %>% 
  tabyl(Role, Sex) %>% 
  adorn_totals(where=c("row", "col")) %>% 
  select(Role, Total, Male, Female) %>%
  # Part 2
  mutate(type = case_when(
    Role %in% c("Mother", "Father")      ~"parent",
    Role %in% c("Proband", "Sibling")    ~"child",
    TRUE                                 ~" "
  )) %>% 
  group_by(type) %>% 
  mutate(total = sum(Total)) %>%
  ungroup() %>%
  select(type, total, Role, Total, Male, Female)
# # A tibble: 5 x 6
#   type   total Role    Total  Male Female
#   <chr>  <dbl> <chr>   <dbl> <dbl>  <dbl>
# 1 parent    6. Father     3.    3.     0.
# 2 parent    6. Mother     3.    0.     3.
# 3 child     4. Proband    3.    2.     1.
# 4 child     4. Sibling    1.    0.     1.
# 5 " "      10. Total     10.    5.     5.

数据

library(tidyverse) library(janitor) sample.df <- read.table(text = "Seq_ID Family Father Mother Sex Role Type 1 SSC02219 11000 0 0 Male Father Parent 2 SSC02217 11000 0 0 Female Mother Parent 3 SSC02254 11000 SSC02219 SSC02217 Male Proband Child 4 SSC02220 11000 SSC02219 SSC02217 Female Sibling Child 5 SSC02184 11001 0 0 Male Father Parent 6 SSC02181 11001 0 0 Female Mother Parent 7 SSC02178 11001 SSC02184 SSC02181 Male Proband Child 8 SSC03092 11002 0 0 Male Father Parent 9 SSC03078 11002 0 0 Female Mother Parent 10 SSC03070 11002 SSC03092 SSC03078 Female Proband Child ", header = TRUE, stringsAsFactors = FALSE) sample.df <- as_tibble(sample.df)

Answer 2

另一种选择可能是使用knitr

library(janitor)
library(tidyverse)
library(kableExtra)
library(knitr)

sample.df %>% 
  tabyl(Role, Sex) %>%
  adorn_totals(where=c("row", "col")) %>%
  # Part 2
  mutate(type=case_when(
    Role %in% c('Father', 'Mother') ~ 'parent',
    Role %in% c('Proband', 'Sibling') ~ 'child',
    TRUE ~ ''
  )) %>%
  group_by(type) %>%
  mutate(total=sum(Total)) %>%
  ungroup() %>%
  kable("html") %>%
  kable_styling(c("striped", "bordered")) %>%
  collapse_rows(columns = c(5,6))

输出为：

示例数据：

sample.df  <- structure(list(Seq_ID = c("SSC02219", "SSC02217", "SSC02254", 
"SSC02220", "SSC02184", "SSC02181", "SSC02178", "SSC03092", "SSC03078", 
"SSC03070"), Family = c(11000L, 11000L, 11000L, 11000L, 11001L, 
11001L, 11001L, 11002L, 11002L, 11002L), Father = c("0", "0", 
"SSC02219", "SSC02219", "0", "0", "SSC02184", "0", "0", "SSC03092"
), Mother = c("0", "0", "SSC02217", "SSC02217", "0", "0", "SSC02181", 
"0", "0", "SSC03078"), Sex = c("Male", "Female", "Male", "Female", 
"Male", "Female", "Male", "Male", "Female", "Female"), Role = c("Father", 
"Mother", "Proband", "Sibling", "Father", "Mother", "Proband", 
"Father", "Mother", "Proband"), Type = c("Parent", "Parent", 
"Child", "Child", "Parent", "Parent", "Child", "Parent", "Parent", 
"Child")), row.names = c("1", "2", "3", "4", "5", "6", "7", "8", 
"9", "10"), class = c("tbl_df", "tbl", "data.frame"))

如何折叠频率表的行以在新列中添加其计数？

2 个答案: