根据不同分组条件创建列
Create a column based on different group conditions
我有一个包含不同列的数据集。看起来像那样
df <- data.frame(PatientID = c("0002" ,"0004", "0005", "0006" ,"0009" ,"0010" ,"0018", "0019" ,"0020" ,"0027", "0039" ,"0041" ,"0042", "0043" ,"0044" ,"0045", "0046", "0047" ,"0048" ,"0049", "0055"),
A = c(987.805 , 977.146 , 790.809 , 964.315 ,1014.020 , 952.311 , 992.967 , 950.797 , 958.975 ,960.712 ,958.117 , 947.465 , 902.852 , 961.417, 985.124 ,994.178 , 930.141 ,1007.790 , 948.848, 1027.110 , 999.414),
B = c(998.988 , 972.606 , 998.680 , 955.037 , 972.941 ,1020.560 , 947.751 ,1029.560 , 955.540 , 911.606 , 964.039 , NA, 988.087 , 902.367 , 959.338 ,1029.050 , 925.162 , 987.374 ,1066.400 ,957.512 , 917.597),
C = c( 975.634 , 987.140 , 961.810 , 929.466 , 978.166, 1005.820 ,925.752 , 969.469 , 943.398 ,936.034, 965.292 , 996.404 , 920.610 , 967.047 ,986.565 , 913.517 , 893.428 , 921.606 , 976.192 , 929.590 ,950.493),
D = c(975.634 , 987.140 , 961.810 , 929.466 , 978.166, 1005.820 , 925.752 , 969.469 ,943.398 , 936.034 , 965.292 , 996.404 , 920.610 , 967.047 , 986.565 , 913.517 , 893.428 , 921.606 , 976.192 , 929.590 , 950.493),
E = c(1006.330, 1028.070 , 975.554 , 954.274 ,1005.910 ,949.969 , 992.820 , 977.048 ,934.407 , 948.913 , 944.578 , 917.564 , 975.301, 961.375 ,955.296 , 961.128 ,998.119 ,1009.110 , 994.891 ,1000.170 ,982.763),
G= c(951.684 , 958.990 , 944.432 , 944.654 , 924.680 , 955.927 , 972.674 , 949.384 ,973.348 , 984.392 , 943.894 , 961.468 , 995.368 , 994.997 , 973.175 , 979.454 , 952.605 , 930.744 , NA, 1015.150 , 956.507), stringsAsFactors = F)
基本上我需要的是创建一个额外的列，将被称为 above threshold，这将是 TRUE/FALSE 基于以下条件：
要成为 TRUE，患者需要在 6 列（A、B、C、D、E 或 G）中的任意 3 列中具有高于阈值的值，阈值是：
- 对于 A 和 B -> 990
- 对于 C 和 D -> 1000
- 对于 E 和 G -> 1005
否则是FALSE
基本上 3 列或更多列需要为 TRUE 才能使最后一列为 TRUE。输出看起来像这样（高于阈值 == TRUE 被涂成绿色）：
我该如何设置？ - 我希望这是清楚的，但如果不是，请问！
非常感谢！
我们创建一个命名的 list（或命名的 vector），遍历 'PatientID' 以外的列，提取具有列名的 list 元素 ( cur_column())，在.names中添加后缀_new创建一个新的逻辑列，然后使用rowSums检查每行TRUE的个数是否大于等于3 创建 'above_threshold'
library(dplyr)
lst1 <- list(A = 990, B = 990, C = 1000, D = 1000, E = 1005, G = 1005)
df %>%
mutate(across(A:G, ~ . > lst1[[cur_column()]],
.names = '{.col}_new'),
above_threshold = rowSums(select(cur_data(), ends_with('new')),
na.rm = TRUE) >=3) %>%
select(names(df), above_threshold)
-输出
PatientID A B C D E G above_threshold
1 0002 987.805 998.988 975.634 975.634 1006.330 951.684 FALSE
2 0004 977.146 972.606 987.140 987.140 1028.070 958.990 FALSE
3 0005 790.809 998.680 961.810 961.810 975.554 944.432 FALSE
4 0006 964.315 955.037 929.466 929.466 954.274 944.654 FALSE
5 0009 1014.020 972.941 978.166 978.166 1005.910 924.680 FALSE
6 0010 952.311 1020.560 1005.820 1005.820 949.969 955.927 TRUE
7 0018 992.967 947.751 925.752 925.752 992.820 972.674 FALSE
8 0019 950.797 1029.560 969.469 969.469 977.048 949.384 FALSE
9 0020 958.975 955.540 943.398 943.398 934.407 973.348 FALSE
10 0027 960.712 911.606 936.034 936.034 948.913 984.392 FALSE
11 0039 958.117 964.039 965.292 965.292 944.578 943.894 FALSE
12 0041 947.465 NA 996.404 996.404 917.564 961.468 FALSE
13 0042 902.852 988.087 920.610 920.610 975.301 995.368 FALSE
14 0043 961.417 902.367 967.047 967.047 961.375 994.997 FALSE
15 0044 985.124 959.338 986.565 986.565 955.296 973.175 FALSE
16 0045 994.178 1029.050 913.517 913.517 961.128 979.454 FALSE
17 0046 930.141 925.162 893.428 893.428 998.119 952.605 FALSE
18 0047 1007.790 987.374 921.606 921.606 1009.110 930.744 FALSE
19 0048 948.848 1066.400 976.192 976.192 994.891 NA FALSE
20 0049 1027.110 957.512 929.590 929.590 1000.170 1015.150 FALSE
21 0055 999.414 917.597 950.493 950.493 982.763 956.507 FALSE
最后，再次在神话般的 akrun 的大力帮助下，这是解决方案：
df %>% mutate(above_treshold = rowSums(cbind(A >= 990, B >= 990, C>=1000, D >= 1000, E >=1005, G >=1005), na.rm = TRUE) >=3)
输出：
PatientID A B C D E G above_treshold
1 0002 987.805 998.988 975.634 975.634 1006.330 951.684 FALSE
2 0004 977.146 972.606 987.140 987.140 1028.070 958.990 FALSE
3 0005 790.809 998.680 961.810 961.810 975.554 944.432 FALSE
4 0006 964.315 955.037 929.466 929.466 954.274 944.654 FALSE
5 0009 1014.020 972.941 978.166 978.166 1005.910 924.680 FALSE
6 0010 952.311 1020.560 1005.820 1005.820 949.969 955.927 TRUE
7 0018 992.967 947.751 925.752 925.752 992.820 972.674 FALSE
8 0019 950.797 1029.560 969.469 969.469 977.048 949.384 FALSE
9 0020 958.975 955.540 943.398 943.398 934.407 973.348 FALSE
10 0027 960.712 911.606 936.034 936.034 948.913 984.392 FALSE
11 0039 958.117 964.039 965.292 965.292 944.578 943.894 FALSE
12 0041 947.465 NA 996.404 996.404 917.564 961.468 FALSE
13 0042 902.852 988.087 920.610 920.610 975.301 995.368 FALSE
14 0043 961.417 902.367 967.047 967.047 961.375 994.997 FALSE
15 0044 985.124 959.338 986.565 986.565 955.296 973.175 FALSE
16 0045 994.178 1029.050 913.517 913.517 961.128 979.454 FALSE
17 0046 930.141 925.162 893.428 893.428 998.119 952.605 FALSE
18 0047 1007.790 987.374 921.606 921.606 1009.110 930.744 FALSE
19 0048 948.848 1066.400 976.192 976.192 994.891 NA FALSE
20 0049 1027.110 957.512 929.590 929.590 1000.170 1015.150 FALSE
21 0055 999.414 917.597 950.493 950.493 982.763 956.507 FALSE
基础
使用@akrun 的想法
standard <- c(A = 990, B = 990, C = 1000, D = 1000, E = 1005, G = 1005)
tmp <- sweep(df[, -1], MARGIN = 2, STATS = standard, FUN = `>=`)
df$res <- apply(tmp, 1, function(x) sum(x, na.rm = TRUE) >= 3)
PatientID A B C D E G res
1 0002 987.805 998.988 975.634 975.634 1006.330 951.684 FALSE
2 0004 977.146 972.606 987.140 987.140 1028.070 958.990 FALSE
3 0005 790.809 998.680 961.810 961.810 975.554 944.432 FALSE
4 0006 964.315 955.037 929.466 929.466 954.274 944.654 FALSE
5 0009 1014.020 972.941 978.166 978.166 1005.910 924.680 FALSE
6 0010 952.311 1020.560 1005.820 1005.820 949.969 955.927 TRUE
7 0018 992.967 947.751 925.752 925.752 992.820 972.674 FALSE
8 0019 950.797 1029.560 969.469 969.469 977.048 949.384 FALSE
9 0020 958.975 955.540 943.398 943.398 934.407 973.348 FALSE
10 0027 960.712 911.606 936.034 936.034 948.913 984.392 FALSE
11 0039 958.117 964.039 965.292 965.292 944.578 943.894 FALSE
12 0041 947.465 NA 996.404 996.404 917.564 961.468 FALSE
13 0042 902.852 988.087 920.610 920.610 975.301 995.368 FALSE
14 0043 961.417 902.367 967.047 967.047 961.375 994.997 FALSE
15 0044 985.124 959.338 986.565 986.565 955.296 973.175 FALSE
16 0045 994.178 1029.050 913.517 913.517 961.128 979.454 FALSE
17 0046 930.141 925.162 893.428 893.428 998.119 952.605 FALSE
18 0047 1007.790 987.374 921.606 921.606 1009.110 930.744 FALSE
19 0048 948.848 1066.400 976.192 976.192 994.891 NA FALSE
20 0049 1027.110 957.512 929.590 929.590 1000.170 1015.150 FALSE
21 0055 999.414 917.597 950.493 950.493 982.763 956.507 FALSE
我有一个包含不同列的数据集。看起来像那样
df <- data.frame(PatientID = c("0002" ,"0004", "0005", "0006" ,"0009" ,"0010" ,"0018", "0019" ,"0020" ,"0027", "0039" ,"0041" ,"0042", "0043" ,"0044" ,"0045", "0046", "0047" ,"0048" ,"0049", "0055"),
A = c(987.805 , 977.146 , 790.809 , 964.315 ,1014.020 , 952.311 , 992.967 , 950.797 , 958.975 ,960.712 ,958.117 , 947.465 , 902.852 , 961.417, 985.124 ,994.178 , 930.141 ,1007.790 , 948.848, 1027.110 , 999.414),
B = c(998.988 , 972.606 , 998.680 , 955.037 , 972.941 ,1020.560 , 947.751 ,1029.560 , 955.540 , 911.606 , 964.039 , NA, 988.087 , 902.367 , 959.338 ,1029.050 , 925.162 , 987.374 ,1066.400 ,957.512 , 917.597),
C = c( 975.634 , 987.140 , 961.810 , 929.466 , 978.166, 1005.820 ,925.752 , 969.469 , 943.398 ,936.034, 965.292 , 996.404 , 920.610 , 967.047 ,986.565 , 913.517 , 893.428 , 921.606 , 976.192 , 929.590 ,950.493),
D = c(975.634 , 987.140 , 961.810 , 929.466 , 978.166, 1005.820 , 925.752 , 969.469 ,943.398 , 936.034 , 965.292 , 996.404 , 920.610 , 967.047 , 986.565 , 913.517 , 893.428 , 921.606 , 976.192 , 929.590 , 950.493),
E = c(1006.330, 1028.070 , 975.554 , 954.274 ,1005.910 ,949.969 , 992.820 , 977.048 ,934.407 , 948.913 , 944.578 , 917.564 , 975.301, 961.375 ,955.296 , 961.128 ,998.119 ,1009.110 , 994.891 ,1000.170 ,982.763),
G= c(951.684 , 958.990 , 944.432 , 944.654 , 924.680 , 955.927 , 972.674 , 949.384 ,973.348 , 984.392 , 943.894 , 961.468 , 995.368 , 994.997 , 973.175 , 979.454 , 952.605 , 930.744 , NA, 1015.150 , 956.507), stringsAsFactors = F)
基本上我需要的是创建一个额外的列，将被称为 above threshold，这将是 TRUE/FALSE 基于以下条件：
要成为 TRUE，患者需要在 6 列（A、B、C、D、E 或 G）中的任意 3 列中具有高于阈值的值，阈值是：
- 对于 A 和 B -> 990
- 对于 C 和 D -> 1000
- 对于 E 和 G -> 1005
否则是FALSE
基本上 3 列或更多列需要为 TRUE 才能使最后一列为 TRUE。输出看起来像这样（高于阈值 == TRUE 被涂成绿色）：
我该如何设置？ - 我希望这是清楚的，但如果不是，请问！
非常感谢！
我们创建一个命名的 list（或命名的 vector），遍历 'PatientID' 以外的列，提取具有列名的 list 元素 ( cur_column())，在.names中添加后缀_new创建一个新的逻辑列，然后使用rowSums检查每行TRUE的个数是否大于等于3 创建 'above_threshold'
library(dplyr)
lst1 <- list(A = 990, B = 990, C = 1000, D = 1000, E = 1005, G = 1005)
df %>%
mutate(across(A:G, ~ . > lst1[[cur_column()]],
.names = '{.col}_new'),
above_threshold = rowSums(select(cur_data(), ends_with('new')),
na.rm = TRUE) >=3) %>%
select(names(df), above_threshold)
-输出
PatientID A B C D E G above_threshold
1 0002 987.805 998.988 975.634 975.634 1006.330 951.684 FALSE
2 0004 977.146 972.606 987.140 987.140 1028.070 958.990 FALSE
3 0005 790.809 998.680 961.810 961.810 975.554 944.432 FALSE
4 0006 964.315 955.037 929.466 929.466 954.274 944.654 FALSE
5 0009 1014.020 972.941 978.166 978.166 1005.910 924.680 FALSE
6 0010 952.311 1020.560 1005.820 1005.820 949.969 955.927 TRUE
7 0018 992.967 947.751 925.752 925.752 992.820 972.674 FALSE
8 0019 950.797 1029.560 969.469 969.469 977.048 949.384 FALSE
9 0020 958.975 955.540 943.398 943.398 934.407 973.348 FALSE
10 0027 960.712 911.606 936.034 936.034 948.913 984.392 FALSE
11 0039 958.117 964.039 965.292 965.292 944.578 943.894 FALSE
12 0041 947.465 NA 996.404 996.404 917.564 961.468 FALSE
13 0042 902.852 988.087 920.610 920.610 975.301 995.368 FALSE
14 0043 961.417 902.367 967.047 967.047 961.375 994.997 FALSE
15 0044 985.124 959.338 986.565 986.565 955.296 973.175 FALSE
16 0045 994.178 1029.050 913.517 913.517 961.128 979.454 FALSE
17 0046 930.141 925.162 893.428 893.428 998.119 952.605 FALSE
18 0047 1007.790 987.374 921.606 921.606 1009.110 930.744 FALSE
19 0048 948.848 1066.400 976.192 976.192 994.891 NA FALSE
20 0049 1027.110 957.512 929.590 929.590 1000.170 1015.150 FALSE
21 0055 999.414 917.597 950.493 950.493 982.763 956.507 FALSE
最后，再次在神话般的 akrun 的大力帮助下，这是解决方案：
df %>% mutate(above_treshold = rowSums(cbind(A >= 990, B >= 990, C>=1000, D >= 1000, E >=1005, G >=1005), na.rm = TRUE) >=3)
输出：
PatientID A B C D E G above_treshold
1 0002 987.805 998.988 975.634 975.634 1006.330 951.684 FALSE
2 0004 977.146 972.606 987.140 987.140 1028.070 958.990 FALSE
3 0005 790.809 998.680 961.810 961.810 975.554 944.432 FALSE
4 0006 964.315 955.037 929.466 929.466 954.274 944.654 FALSE
5 0009 1014.020 972.941 978.166 978.166 1005.910 924.680 FALSE
6 0010 952.311 1020.560 1005.820 1005.820 949.969 955.927 TRUE
7 0018 992.967 947.751 925.752 925.752 992.820 972.674 FALSE
8 0019 950.797 1029.560 969.469 969.469 977.048 949.384 FALSE
9 0020 958.975 955.540 943.398 943.398 934.407 973.348 FALSE
10 0027 960.712 911.606 936.034 936.034 948.913 984.392 FALSE
11 0039 958.117 964.039 965.292 965.292 944.578 943.894 FALSE
12 0041 947.465 NA 996.404 996.404 917.564 961.468 FALSE
13 0042 902.852 988.087 920.610 920.610 975.301 995.368 FALSE
14 0043 961.417 902.367 967.047 967.047 961.375 994.997 FALSE
15 0044 985.124 959.338 986.565 986.565 955.296 973.175 FALSE
16 0045 994.178 1029.050 913.517 913.517 961.128 979.454 FALSE
17 0046 930.141 925.162 893.428 893.428 998.119 952.605 FALSE
18 0047 1007.790 987.374 921.606 921.606 1009.110 930.744 FALSE
19 0048 948.848 1066.400 976.192 976.192 994.891 NA FALSE
20 0049 1027.110 957.512 929.590 929.590 1000.170 1015.150 FALSE
21 0055 999.414 917.597 950.493 950.493 982.763 956.507 FALSE
基础
使用@akrun 的想法
standard <- c(A = 990, B = 990, C = 1000, D = 1000, E = 1005, G = 1005)
tmp <- sweep(df[, -1], MARGIN = 2, STATS = standard, FUN = `>=`)
df$res <- apply(tmp, 1, function(x) sum(x, na.rm = TRUE) >= 3)
PatientID A B C D E G res
1 0002 987.805 998.988 975.634 975.634 1006.330 951.684 FALSE
2 0004 977.146 972.606 987.140 987.140 1028.070 958.990 FALSE
3 0005 790.809 998.680 961.810 961.810 975.554 944.432 FALSE
4 0006 964.315 955.037 929.466 929.466 954.274 944.654 FALSE
5 0009 1014.020 972.941 978.166 978.166 1005.910 924.680 FALSE
6 0010 952.311 1020.560 1005.820 1005.820 949.969 955.927 TRUE
7 0018 992.967 947.751 925.752 925.752 992.820 972.674 FALSE
8 0019 950.797 1029.560 969.469 969.469 977.048 949.384 FALSE
9 0020 958.975 955.540 943.398 943.398 934.407 973.348 FALSE
10 0027 960.712 911.606 936.034 936.034 948.913 984.392 FALSE
11 0039 958.117 964.039 965.292 965.292 944.578 943.894 FALSE
12 0041 947.465 NA 996.404 996.404 917.564 961.468 FALSE
13 0042 902.852 988.087 920.610 920.610 975.301 995.368 FALSE
14 0043 961.417 902.367 967.047 967.047 961.375 994.997 FALSE
15 0044 985.124 959.338 986.565 986.565 955.296 973.175 FALSE
16 0045 994.178 1029.050 913.517 913.517 961.128 979.454 FALSE
17 0046 930.141 925.162 893.428 893.428 998.119 952.605 FALSE
18 0047 1007.790 987.374 921.606 921.606 1009.110 930.744 FALSE
19 0048 948.848 1066.400 976.192 976.192 994.891 NA FALSE
20 0049 1027.110 957.512 929.590 929.590 1000.170 1015.150 FALSE
21 0055 999.414 917.597 950.493 950.493 982.763 956.507 FALSE