R：根据另一个数据框过滤数据框
R: Filter a dataframe based on another dataframe
我想根据 pf df 对 e df 进行子集化。这两个数据框共享相同的行名。我的代码添加了一个“row_names”列，这是我不想要的。
library(dplyr)
e <- e[, !row.names(e) %in% row.names(pf)]
> dput(e[1:10, 1:10])
structure(list(row_names = c("2315374", "2315376", "2315587",
"2315588", "2315591", "2315595", "2315598", "2315603", "2315604",
"2315640"), JHU_113_2.CEL = c(6.28274, 5.81678, 8.88557, 6.2865,
6.97515, 5.94179, 4.9942, 7.67845, 6.20952, 5.85307), JHU_144.CEL = c(6.79161,
5.71165, 8.95699, 6.6675, 6.61705, 5.39178, 5.56888, 7.90005,
6.59687, 6.07303), JHU_173.CEL = c(6.11265, 6.02794, 8.36898,
6.07503, 6.51994, 5.09497, 5.57912, 7.47594, 6.14608, 6.41875
), JHU_176R.CEL = c(6.13997, 5.37082, 8.28993, 6.76625, 6.74982,
4.96199, 5.4396, 6.75087, 5.70518, 6.07282), JHU_182.CEL = c(6.68056,
5.95527, 8.41361, 6.19819, 6.60917, 2.96431, 5.19249, 7.62805,
6.49572, 6.28283), JHU_186.CEL = c(6.48156, 5.75999, 8.6498,
6.8426, 6.55182, 4.95204, 5.87991, 8.00069, 6.12622, 6.13699),
JHU_187.CEL = c(6.45415, 5.87863, 8.74305, 6.13916, 6.6224,
5.00979, 5.6054, 7.34296, 6.2369, 6.16377), JHU_188.CEL = c(6.04542,
5.5483, 8.31915, 6.40219, 6.44394, 4.06493, 5.09513, 6.81338,
6.39569, 6.48616), JHU_203.CEL = c(5.99176, 6.35571, 8.43548,
6.45059, 5.76592, 5.38048, 5.43618, 7.52014, 6.70869, 6.34162
)), row.names = c(NA, 10L), class = "data.frame")
> dput(pf[1:10, 1:10])
structure(list(JHU_113_2.CEL = c(0.0073, 0.04079, 0.02234, 0.08252,
0.00127, 0.00256, 0.04265, 0.02244, 0.02615, 0.04563), JHU_144.CEL = c(0.00293,
0.01525, 0.00914, 0.03644, 0.00162, 0.01274, 0.00798, 0.00955,
0.00732, 0.02698), JHU_173.CEL = c(0.03034, 0.0309, 0.05024,
0.09374, 0.00548, 0.04754, 0.03491, 0.03399, 0.02661, 0.01546
), JHU_176R.CEL = c(0.02571, 0.08493, 0.06918, 0.0337, 0.00945,
0.03185, 0.03182, 0.15281, 0.04117, 0.03758), JHU_182.CEL = c(0.00436,
0.01303, 0.0521, 0.04909, 0.00341, 0.45759, 0.02033, 0.01911,
0.01037, 0.01974), JHU_186.CEL = c(0.00788, 0.03559, 0.02822,
0.02419, 0.00383, 0.02401, 0.00727, 0.0126, 0.01663, 0.02291),
JHU_187.CEL = c(0.0052, 0.02163, 0.02401, 0.07845, 0.0028,
0.01582, 0.008, 0.0432, 0.01312, 0.03307), JHU_188.CEL = c(0.01704,
0.01586, 0.06648, 0.09027, 0.00882, 0.21313, 0.03859, 0.1424,
0.01815, 0.01586), JHU_203.CEL = c(0.03273, 0.04264, 0.06823,
0.09449, 0.06553, 0.05349, 0.06481, 0.07518, 0.02777, 0.07511
), JHU_205.CEL = c(0.0672, 0.05689, 0.1278, 0.12526, 0.03327,
0.10091, 0.42112, 0.18431, 0.08595, 0.01157)), row.names = c(2315374L,
2315376L, 2315587L, 2315588L, 2315591L, 2315595L, 2315598L, 2315603L,
2315604L, 2315640L), class = "data.frame")
如果您只想保留 e 中出现在 pf 中的行名（或者不出现，则使用 !rownames(e)），那么您可以 filter 在行名上：
library(tidyverse)
e %>%
filter(rownames(e) %in% rownames(pf))
另一种可能性是为两个数据框创建一个行名列。然后，我们可以对行名执行 semi_join（即 rn）。然后，将 rn 列转换回行名。
library(tidyverse)
list(e, pf) %>%
map(~ .x %>%
as.data.frame %>%
rownames_to_column('rn')) %>%
reduce(full_join, by = 'rn') %>%
column_to_rownames('rn')
输出
JHU_113_2.CEL JHU_144.CEL JHU_173.CEL JHU_176R.CEL JHU_182.CEL JHU_186.CEL JHU_187.CEL JHU_188.CEL JHU_203.CEL
2315374 6.28274 6.79161 6.11265 6.13997 6.68056 6.48156 6.45415 6.04542 5.99176
2315376 5.81678 5.71165 6.02794 5.37082 5.95527 5.75999 5.87863 5.54830 6.35571
2315587 8.88557 8.95699 8.36898 8.28993 8.41361 8.64980 8.74305 8.31915 8.43548
2315588 6.28650 6.66750 6.07503 6.76625 6.19819 6.84260 6.13916 6.40219 6.45059
2315591 6.97515 6.61705 6.51994 6.74982 6.60917 6.55182 6.62240 6.44394 5.76592
2315595 5.94179 5.39178 5.09497 4.96199 2.96431 4.95204 5.00979 4.06493 5.38048
2315598 4.99420 5.56888 5.57912 5.43960 5.19249 5.87991 5.60540 5.09513 5.43618
2315603 7.67845 7.90005 7.47594 6.75087 7.62805 8.00069 7.34296 6.81338 7.52014
2315604 6.20952 6.59687 6.14608 5.70518 6.49572 6.12622 6.23690 6.39569 6.70869
2315640 5.85307 6.07303 6.41875 6.07282 6.28283 6.13699 6.16377 6.48616 6.34162
我想根据 pf df 对 e df 进行子集化。这两个数据框共享相同的行名。我的代码添加了一个“row_names”列，这是我不想要的。
library(dplyr)
e <- e[, !row.names(e) %in% row.names(pf)]
> dput(e[1:10, 1:10])
structure(list(row_names = c("2315374", "2315376", "2315587",
"2315588", "2315591", "2315595", "2315598", "2315603", "2315604",
"2315640"), JHU_113_2.CEL = c(6.28274, 5.81678, 8.88557, 6.2865,
6.97515, 5.94179, 4.9942, 7.67845, 6.20952, 5.85307), JHU_144.CEL = c(6.79161,
5.71165, 8.95699, 6.6675, 6.61705, 5.39178, 5.56888, 7.90005,
6.59687, 6.07303), JHU_173.CEL = c(6.11265, 6.02794, 8.36898,
6.07503, 6.51994, 5.09497, 5.57912, 7.47594, 6.14608, 6.41875
), JHU_176R.CEL = c(6.13997, 5.37082, 8.28993, 6.76625, 6.74982,
4.96199, 5.4396, 6.75087, 5.70518, 6.07282), JHU_182.CEL = c(6.68056,
5.95527, 8.41361, 6.19819, 6.60917, 2.96431, 5.19249, 7.62805,
6.49572, 6.28283), JHU_186.CEL = c(6.48156, 5.75999, 8.6498,
6.8426, 6.55182, 4.95204, 5.87991, 8.00069, 6.12622, 6.13699),
JHU_187.CEL = c(6.45415, 5.87863, 8.74305, 6.13916, 6.6224,
5.00979, 5.6054, 7.34296, 6.2369, 6.16377), JHU_188.CEL = c(6.04542,
5.5483, 8.31915, 6.40219, 6.44394, 4.06493, 5.09513, 6.81338,
6.39569, 6.48616), JHU_203.CEL = c(5.99176, 6.35571, 8.43548,
6.45059, 5.76592, 5.38048, 5.43618, 7.52014, 6.70869, 6.34162
)), row.names = c(NA, 10L), class = "data.frame")
> dput(pf[1:10, 1:10])
structure(list(JHU_113_2.CEL = c(0.0073, 0.04079, 0.02234, 0.08252,
0.00127, 0.00256, 0.04265, 0.02244, 0.02615, 0.04563), JHU_144.CEL = c(0.00293,
0.01525, 0.00914, 0.03644, 0.00162, 0.01274, 0.00798, 0.00955,
0.00732, 0.02698), JHU_173.CEL = c(0.03034, 0.0309, 0.05024,
0.09374, 0.00548, 0.04754, 0.03491, 0.03399, 0.02661, 0.01546
), JHU_176R.CEL = c(0.02571, 0.08493, 0.06918, 0.0337, 0.00945,
0.03185, 0.03182, 0.15281, 0.04117, 0.03758), JHU_182.CEL = c(0.00436,
0.01303, 0.0521, 0.04909, 0.00341, 0.45759, 0.02033, 0.01911,
0.01037, 0.01974), JHU_186.CEL = c(0.00788, 0.03559, 0.02822,
0.02419, 0.00383, 0.02401, 0.00727, 0.0126, 0.01663, 0.02291),
JHU_187.CEL = c(0.0052, 0.02163, 0.02401, 0.07845, 0.0028,
0.01582, 0.008, 0.0432, 0.01312, 0.03307), JHU_188.CEL = c(0.01704,
0.01586, 0.06648, 0.09027, 0.00882, 0.21313, 0.03859, 0.1424,
0.01815, 0.01586), JHU_203.CEL = c(0.03273, 0.04264, 0.06823,
0.09449, 0.06553, 0.05349, 0.06481, 0.07518, 0.02777, 0.07511
), JHU_205.CEL = c(0.0672, 0.05689, 0.1278, 0.12526, 0.03327,
0.10091, 0.42112, 0.18431, 0.08595, 0.01157)), row.names = c(2315374L,
2315376L, 2315587L, 2315588L, 2315591L, 2315595L, 2315598L, 2315603L,
2315604L, 2315640L), class = "data.frame")
如果您只想保留 e 中出现在 pf 中的行名（或者不出现，则使用 !rownames(e)），那么您可以 filter 在行名上：
library(tidyverse)
e %>%
filter(rownames(e) %in% rownames(pf))
另一种可能性是为两个数据框创建一个行名列。然后，我们可以对行名执行 semi_join（即 rn）。然后，将 rn 列转换回行名。
library(tidyverse)
list(e, pf) %>%
map(~ .x %>%
as.data.frame %>%
rownames_to_column('rn')) %>%
reduce(full_join, by = 'rn') %>%
column_to_rownames('rn')
输出
JHU_113_2.CEL JHU_144.CEL JHU_173.CEL JHU_176R.CEL JHU_182.CEL JHU_186.CEL JHU_187.CEL JHU_188.CEL JHU_203.CEL
2315374 6.28274 6.79161 6.11265 6.13997 6.68056 6.48156 6.45415 6.04542 5.99176
2315376 5.81678 5.71165 6.02794 5.37082 5.95527 5.75999 5.87863 5.54830 6.35571
2315587 8.88557 8.95699 8.36898 8.28993 8.41361 8.64980 8.74305 8.31915 8.43548
2315588 6.28650 6.66750 6.07503 6.76625 6.19819 6.84260 6.13916 6.40219 6.45059
2315591 6.97515 6.61705 6.51994 6.74982 6.60917 6.55182 6.62240 6.44394 5.76592
2315595 5.94179 5.39178 5.09497 4.96199 2.96431 4.95204 5.00979 4.06493 5.38048
2315598 4.99420 5.56888 5.57912 5.43960 5.19249 5.87991 5.60540 5.09513 5.43618
2315603 7.67845 7.90005 7.47594 6.75087 7.62805 8.00069 7.34296 6.81338 7.52014
2315604 6.20952 6.59687 6.14608 5.70518 6.49572 6.12622 6.23690 6.39569 6.70869
2315640 5.85307 6.07303 6.41875 6.07282 6.28283 6.13699 6.16377 6.48616 6.34162