更新“tibble”版本时“pivot_longer()”参数出错
Error in `pivot_longer()` argument when updating `tibble` versions
上下文
我们想管理从宽格式 dat_1 到长格式 dat_2 的数据。为了执行它，我们考虑使用 tidyr::pivot_longer() 和参数 names_pattern = '(.+)_(.+)'。它允许我们收集数据，如框中所示，从 input format 到 output format。
输入数据
> dat_1
original_id timepoint msp3_mfi msp3_dil pf_aarp_mfi pf_aarp_dil
<chr> <chr> <dbl> <int> <dbl> <int>
1 id_005 C_0 10.5 400 22.2 400
2 id_005 D10 8.5 400 10.25 400
3 id_005 D13 11 400 10.2 400
4 id_005 D28 8 400 9.75 400
5 id_005 D60 7 400 0.30 400
使用了R代码
dat_2 <- dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', 'antigen_dil'),
names_pattern = '(.+)_(.+)',
values_to = c('mfi', 'dil'))
期望的输出数据
> dat_2
original_id timepoint antigen antigen_dil mfi dil
<chr> <chr> <chr> <chr> <dbl> <int>
1 id_005 C_0 msp3 mfi 10.5 NA
2 id_005 C_0 msp3 dil NA 400
3 id_005 C_0 pf_aarp mfi 22.2 NA
4 id_005 C_0 pf_aarp dil NA 400
5 id_005 D10 msp3 mfi 8.5 NA
6 id_005 D10 msp3 dil NA 400
7 id_005 D10 pf_aarp mfi 10.25 NA
8 id_005 D10 pf_aarp dil NA 400
9 id_005 D13 msp3 mfi 11 NA
10 id_005 D13 msp3 dil NA 400
11 id_005 D13 pf_aarp mfi 10.2 NA
12 id_005 D13 pf_aarp dil NA 400
13 id_005 D28 msp3 mfi 8 NA
14 id_005 D28 msp3 dil NA 400
15 id_005 D28 pf_aarp mfi 9.75 NA
16 id_005 D28 pf_aarp dil NA 400
17 id_005 D60 msp3 mfi 7 NA
18 id_005 D60 msp3 dil NA 400
19 id_005 D60 pf_aarp mfi 0.30 NA
20 id_005 D60 pf_aarp dil NA 400
错误
但是，当我们将 de R（版本 R 3.6.3）会话从 tibble 2.1.3 更新到 tibble. 3.0.1 时，我们会收到以下错误。
Error: Assigned data `values_to` must be compatible with existing data.
x Existing data has 4 rows.
x Assigned data has 2 rows.
i Only vectors of size 1 are recycled.
Run `rlang::last_error()` to see where the error occurred.
问题
知道为什么我们在更新 tibble 软件包版本时在 name_pattern 中出现此错误吗？
提前致谢
虽然我不知道为什么会发生变化，但我猜想，从本质上讲，这个错误与你的 dat_2 tibble 没有 tidy。重复将 "mfi" 和 "dil" 之间的差异作为其自己的列 antigen_dil 和两个单独的列 mfi 和 dil.
根据您的数据的含义，可以轻松使用 pivot_longer 的两种格式是：
dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', '.value'),
names_pattern = '(.+)_(.+)'
)
#> # A tibble: 4 x 5
#> original_id timepoint antigen mfi dil
#> <chr> <chr> <chr> <dbl> <dbl>
#> 1 id_005 C_0 msp3 10.5 400
#> 2 id_005 C_0 pf_aarp 22.2 400
#> 3 id_005 D10 msp3 8.5 400
#> 4 id_005 D10 pf_aarp 10.2 400
或
dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', 'antigen_dil'),
names_pattern = '(.+)_(.+)'
)
#> # A tibble: 8 x 5
#> original_id timepoint antigen antigen_dil value
#> <chr> <chr> <chr> <chr> <dbl>
#> 1 id_005 C_0 msp3 mfi 10.5
#> 2 id_005 C_0 msp3 dil 400
#> 3 id_005 C_0 pf_aarp mfi 22.2
#> 4 id_005 C_0 pf_aarp dil 400
#> 5 id_005 D10 msp3 mfi 8.5
#> 6 id_005 D10 msp3 dil 400
#> 7 id_005 D10 pf_aarp mfi 10.2
#> 8 id_005 D10 pf_aarp dil 400
如果您确实需要您所描述格式的小标题，您可以使用：
dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', 'antigen_dil'),
names_pattern = '(.+)_(.+)'
) %>%
mutate(
mfi = if_else(antigen_dil == "mfi", value, NA_real_),
dil = if_else(antigen_dil == "dil", value, NA_real_)
) %>%
select(-value)
#> # A tibble: 8 x 6
#> original_id timepoint antigen antigen_dil mfi dil
#> <chr> <chr> <chr> <chr> <dbl> <dbl>
#> 1 id_005 C_0 msp3 mfi 10.5 NA
#> 2 id_005 C_0 msp3 dil NA 400
#> 3 id_005 C_0 pf_aarp mfi 22.2 NA
#> 4 id_005 C_0 pf_aarp dil NA 400
#> 5 id_005 D10 msp3 mfi 8.5 NA
#> 6 id_005 D10 msp3 dil NA 400
#> 7 id_005 D10 pf_aarp mfi 10.2 NA
#> 8 id_005 D10 pf_aarp dil NA 400
以上代码片段使用以下内容创建 dat_1：
library(tidyverse)
dat_1 <-
tribble(
~original_id, ~timepoint, ~msp3_mfi, ~msp3_dil, ~pf_aarp_mfi, ~pf_aarp_dil,
"id_005", "C_0", 10.5, 400, 22.2, 400,
"id_005", "D10", 8.5, 400, 10.25, 400
)
dat_1
#> # A tibble: 2 x 6
#> original_id timepoint msp3_mfi msp3_dil pf_aarp_mfi pf_aarp_dil
#> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 id_005 C_0 10.5 400 22.2 400
#> 2 id_005 D10 8.5 400 10.2 400
上下文
我们想管理从宽格式 dat_1 到长格式 dat_2 的数据。为了执行它，我们考虑使用 tidyr::pivot_longer() 和参数 names_pattern = '(.+)_(.+)'。它允许我们收集数据，如框中所示，从 input format 到 output format。
输入数据
> dat_1
original_id timepoint msp3_mfi msp3_dil pf_aarp_mfi pf_aarp_dil
<chr> <chr> <dbl> <int> <dbl> <int>
1 id_005 C_0 10.5 400 22.2 400
2 id_005 D10 8.5 400 10.25 400
3 id_005 D13 11 400 10.2 400
4 id_005 D28 8 400 9.75 400
5 id_005 D60 7 400 0.30 400
使用了R代码
dat_2 <- dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', 'antigen_dil'),
names_pattern = '(.+)_(.+)',
values_to = c('mfi', 'dil'))
期望的输出数据
> dat_2
original_id timepoint antigen antigen_dil mfi dil
<chr> <chr> <chr> <chr> <dbl> <int>
1 id_005 C_0 msp3 mfi 10.5 NA
2 id_005 C_0 msp3 dil NA 400
3 id_005 C_0 pf_aarp mfi 22.2 NA
4 id_005 C_0 pf_aarp dil NA 400
5 id_005 D10 msp3 mfi 8.5 NA
6 id_005 D10 msp3 dil NA 400
7 id_005 D10 pf_aarp mfi 10.25 NA
8 id_005 D10 pf_aarp dil NA 400
9 id_005 D13 msp3 mfi 11 NA
10 id_005 D13 msp3 dil NA 400
11 id_005 D13 pf_aarp mfi 10.2 NA
12 id_005 D13 pf_aarp dil NA 400
13 id_005 D28 msp3 mfi 8 NA
14 id_005 D28 msp3 dil NA 400
15 id_005 D28 pf_aarp mfi 9.75 NA
16 id_005 D28 pf_aarp dil NA 400
17 id_005 D60 msp3 mfi 7 NA
18 id_005 D60 msp3 dil NA 400
19 id_005 D60 pf_aarp mfi 0.30 NA
20 id_005 D60 pf_aarp dil NA 400
错误
但是，当我们将 de R（版本 R 3.6.3）会话从 tibble 2.1.3 更新到 tibble. 3.0.1 时，我们会收到以下错误。
Error: Assigned data `values_to` must be compatible with existing data.
x Existing data has 4 rows.
x Assigned data has 2 rows.
i Only vectors of size 1 are recycled.
Run `rlang::last_error()` to see where the error occurred.
问题
知道为什么我们在更新 tibble 软件包版本时在 name_pattern 中出现此错误吗？
提前致谢
虽然我不知道为什么会发生变化，但我猜想，从本质上讲，这个错误与你的 dat_2 tibble 没有 tidy。重复将 "mfi" 和 "dil" 之间的差异作为其自己的列 antigen_dil 和两个单独的列 mfi 和 dil.
根据您的数据的含义，可以轻松使用 pivot_longer 的两种格式是：
dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', '.value'),
names_pattern = '(.+)_(.+)'
)
#> # A tibble: 4 x 5
#> original_id timepoint antigen mfi dil
#> <chr> <chr> <chr> <dbl> <dbl>
#> 1 id_005 C_0 msp3 10.5 400
#> 2 id_005 C_0 pf_aarp 22.2 400
#> 3 id_005 D10 msp3 8.5 400
#> 4 id_005 D10 pf_aarp 10.2 400
或
dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', 'antigen_dil'),
names_pattern = '(.+)_(.+)'
)
#> # A tibble: 8 x 5
#> original_id timepoint antigen antigen_dil value
#> <chr> <chr> <chr> <chr> <dbl>
#> 1 id_005 C_0 msp3 mfi 10.5
#> 2 id_005 C_0 msp3 dil 400
#> 3 id_005 C_0 pf_aarp mfi 22.2
#> 4 id_005 C_0 pf_aarp dil 400
#> 5 id_005 D10 msp3 mfi 8.5
#> 6 id_005 D10 msp3 dil 400
#> 7 id_005 D10 pf_aarp mfi 10.2
#> 8 id_005 D10 pf_aarp dil 400
如果您确实需要您所描述格式的小标题，您可以使用：
dat_1 %>%
pivot_longer(
cols = msp3_mfi:pf_aarp_dil,
names_to = c('antigen', 'antigen_dil'),
names_pattern = '(.+)_(.+)'
) %>%
mutate(
mfi = if_else(antigen_dil == "mfi", value, NA_real_),
dil = if_else(antigen_dil == "dil", value, NA_real_)
) %>%
select(-value)
#> # A tibble: 8 x 6
#> original_id timepoint antigen antigen_dil mfi dil
#> <chr> <chr> <chr> <chr> <dbl> <dbl>
#> 1 id_005 C_0 msp3 mfi 10.5 NA
#> 2 id_005 C_0 msp3 dil NA 400
#> 3 id_005 C_0 pf_aarp mfi 22.2 NA
#> 4 id_005 C_0 pf_aarp dil NA 400
#> 5 id_005 D10 msp3 mfi 8.5 NA
#> 6 id_005 D10 msp3 dil NA 400
#> 7 id_005 D10 pf_aarp mfi 10.2 NA
#> 8 id_005 D10 pf_aarp dil NA 400
以上代码片段使用以下内容创建 dat_1：
library(tidyverse)
dat_1 <-
tribble(
~original_id, ~timepoint, ~msp3_mfi, ~msp3_dil, ~pf_aarp_mfi, ~pf_aarp_dil,
"id_005", "C_0", 10.5, 400, 22.2, 400,
"id_005", "D10", 8.5, 400, 10.25, 400
)
dat_1
#> # A tibble: 2 x 6
#> original_id timepoint msp3_mfi msp3_dil pf_aarp_mfi pf_aarp_dil
#> <chr> <chr> <dbl> <dbl> <dbl> <dbl>
#> 1 id_005 C_0 10.5 400 22.2 400
#> 2 id_005 D10 8.5 400 10.2 400