我如何编写代码在一个分组内搜索多个列 – R

huangapple
go评论96阅读模式
英文:

How can I code to search across multiple columns within a group - R

问题

我有癌症数据，每个患者有1-4个测量值。一些测量进行了细胞学检查，其他一些进行了病理学检查，有些两者都有。

  1. 库（dplyr）
  2. 库（鹈鹕）
  4. 数据<-简表(
  5.   ~record_number, ~tool, ~cytology, ~pathology,
  6.   114, "forceps", "Indeterminate", NA,
  7.   114, "needle", "Non-Malignant", "Malignant",
  8.   114, "lavage", NA, "Indeterminate",
  9.   115, "forceps", NA, "Non-Malignant",
  10.   115, "needle", NA, "Malignant"
  11. )

我想创建一个恶性变量（0/1），如果对于给定主题（record_number）的任何样本（行），在任何列（cytology、pathology）中出现"Malignant"。

任何想法都受到赞赏！

  1. 期望的<-简表(
  2.   ~record_number, ~tool, ~cytology, ~pathology, ~ Malignant,
  3.   114, "forceps", "Indeterminate", NA, 1,
  4.   114, "needle", "Non-Malignant", "Malignant", 1,
  5.   114, "lavage", NA, "Indeterminate", 1, 
  6.   115, "forceps", NA, "Non-Malignant", 1, 
  7.   115, "needle", NA, "Malignant", 1,
  8. )

我想它将以group_by（record_number）开始...但接下来呢？

  1. 期望的<-数据 %>%
  2.   group_by(record_number) %>%
  3.   ...?
英文:

I have cancer data and each patient had 1-4 measurements. Some measurements had cytology done, others had pathology done, some had both.

  1. library(dplyr)
  2. library(tibble)
  4. data<-tribble(
  5.   ~record_number, ~tool, ~cytology, ~pathology,
  6.   114, "forceps", "Indeterminate", NA,
  7.   114, "needle", "Non-Malignant", "Malignant",
  8.   114, "lavage", NA, "Indeterminate",
  9.   115, "forceps", NA, "Non-Malignant",
  10.   115, "needle", NA, "Malignant"
  11. )

I'd like to create a Malignancy variable (0/1) if "Malignant" occurs for any of the samples (rows) for a given subject (record_number), in either of the columns (cytology, pathology).

Any ideas are appreciated!

  1. desired<-tribble(
  2.   ~record_number, ~tool, ~cytology, ~pathology, ~ Malignant,
  3.   114, "forceps", "Indeterminate", NA, 1,
  4.   114, "needle", "Non-Malignant", "Malignant", 1,
  5.   114, "lavage", NA, "Indeterminate", 1, 
  6.   115, "forceps", NA, "Non-Malignant", 1, 
  7.   115, "needle", NA, "Malignant", 1,
  8. )

I'm thinking it will start with group_by(record_number)...but then what?

  1. desired<-data %>%
  2.   group_by(record_number) %>%
  3.   ...?

答案1

得分: 2

你可以这样做：

  1. library(tidyverse)
  2. data %>%
  3.   mutate(Malignant = as.numeric(any(c(cytology, pathology) == 'Malignant', na.rm = TRUE)), .by = record_number)
  4.   
  5. # 一个 tibble: 5 x 5
  6.   record_number tool    cytology      pathology     Malignant
  7.           <dbl> <chr>   <chr>         <chr>             <dbl>
  8. 1           114 forceps Indeterminate <NA>                  1
  9. 2           114 needle  Non-Malignant Malignant             1
  10. 3           114 lavage  <NA>          Indeterminate         1
  11. 4           115 forceps <NA>          Non-Malignant         1
  12. 5           115 needle  <NA>          Malignant             1
英文:

You could do:

  1. library(tidyverse)
  2. data %&gt;%
  3.   mutate(Malignant = as.numeric(any(c(cytology, pathology) == &#39;Malignant&#39;, na.rm = TRUE)), .by = record_number)
  5. # A tibble: 5 x 5
  6.   record_number tool    cytology      pathology     Malignant
  7.           &lt;dbl&gt; &lt;chr&gt;   &lt;chr&gt;         &lt;chr&gt;             &lt;dbl&gt;
  8. 1           114 forceps Indeterminate &lt;NA&gt;                  1
  9. 2           114 needle  Non-Malignant Malignant             1
  10. 3           114 lavage  &lt;NA&gt;          Indeterminate         1
  11. 4           115 forceps &lt;NA&gt;          Non-Malignant         1
  12. 5           115 needle  &lt;NA&gt;          Malignant             1

答案2

得分: 1

我们可以使用 ifelse 和 any 进行条件判断：

  1. library(dplyr) #&gt; 1.1.0
  3. data %&gt;%
  4.   mutate(Malignant = ifelse(any(cytology == &quot;Malignant&quot; | pathology == &quot;Malignant&quot;, na.rm = TRUE), 1, 0), .by=record_number)
  6.   record_number tool    cytology      pathology     Malignant
  7.           &lt;dbl&gt; &lt;chr&gt;   &lt;chr&gt;         &lt;chr&gt;             &lt;dbl&gt;
  8. 1           114 forceps Indeterminate NA                    1
  9. 2           114 needle  Non-Malignant Malignant             1
  10. 3           114 lavage  NA            Indeterminate         1
  11. 4           115 forceps NA            Non-Malignant         1
  12. 5           115 needle  NA            Malignant             1
英文:

We can ifelse with any:

  1. library(dplyr) #&gt; 1.1.0
  3. data %&gt;%
  4.   mutate(Malignant = ifelse(any(cytology == &quot;Malignant&quot; | pathology == &quot;Malignant&quot;, na.rm = TRUE), 1, 0), .by=record_number)
  6.   record_number tool    cytology      pathology     Malignant
  7.           &lt;dbl&gt; &lt;chr&gt;   &lt;chr&gt;         &lt;chr&gt;             &lt;dbl&gt;
  8. 1           114 forceps Indeterminate NA                    1
  9. 2           114 needle  Non-Malignant Malignant             1
  10. 3           114 lavage  NA            Indeterminate         1
  11. 4           115 forceps NA            Non-Malignant         1
  12. 5           115 needle  NA            Malignant             1

答案3

得分: 1

你也可以使用%in%运算符，如果存在NA值，它不会返回NA：

  1. data |&gt;
  2.   mutate(malignancy = 1*("Malignant" %in% c(cytology, pathology)), .by = record_number)

输出

  1.   record_number tool    cytology      pathology     malignancy
  2. 1           114 forceps Indeterminate NA                     1
  3. 2           114 needle  Non-Malignant Malignant              1
  4. 3           114 lavage  NA            Indeterminate          1
  5. 4           115 forceps NA            Non-Malignant          1
  6. 5           115 needle  NA            Malignant              1
英文:

You can use the %in% operator for this too which will not return NA if there are NAs:

  1. data |&gt;
  2.   mutate(malignancy = 1*(&quot;Malignant&quot; %in% c(cytology, pathology)), .by = record_number)

Output

  1.   record_number tool    cytology      pathology     malignancy
  2. 1           114 forceps Indeterminate NA                     1
  3. 2           114 needle  Non-Malignant Malignant              1
  4. 3           114 lavage  NA            Indeterminate          1
  5. 4           115 forceps NA            Non-Malignant          1
  6. 5           115 needle  NA            Malignant              1

huangapple
  • 本文由 发表于 2023年5月24日 22:27:44
  • 转载请务必保留本文链接：https://go.coder-hub.com/76324621.html
  • dplyr
  • r
匿名

发表评论

匿名网友

:?: :razz: :sad: :evil: :!: :smile: :oops: :grin: :eek: :shock: :???: :cool: :lol: :mad: :twisted: :roll: :wink: :idea: :arrow: :neutral: :cry: :mrgreen:

确定