我有两个数据框，一个是原始数据：
df <- structure(list(day = c(2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2), Subject = c(1, 1, 1, 1, 1, 1, 3, 3,
3, 3, 3, 3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 12, 12, 12, 12,
12, 12, 13, 13, 13, 13, 13, 13, 17, 17, 17, 17, 17, 17), TimePoint = c(1,
5, 9, 13, 17, 21, 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23,
1, 5, 9, 15, 17, 21, 1, 5, 9, 13, 17, 21, 1, 5, 9, 13, 17, 21,
1, 5, 9, 13, 17, 21), C4b = c(489.1, 3757.5, 4013.7, 963.4, 668.8,
2135.4, 1793.5, 2485.9, 1337.2, 1200.1, 1854.5, 1163.8, 1050.4,
1514.7, 1134, 1283.9, 1883.2, 1160, 441.4, 318.4, 261.8, 392,
380.9, 312, 1731.1, 1482.9, 1635.6, 1144.3, 1455.3, 1427.1, 1202.6,
1212.7, 1294.8, 1406.6, 2153.3, 1135, 1104.5, 1129.3, 737.4,
1010.9, 934.3, 704.2), `Coagulation Factor XI` = c(2149.4, 2055.8,
2256.3, 2179, 1859.5, 2158.4, 1877.4, 1800.7, 1838.7, 1899.9,
1867, 1827.9, 1818.7, 1847, 1807.7, 1737, 2394.9, 1726.2, 1756.3,
1746.3, 1744.5, 1573.8, 1765.5, 1954.1, 1522.4, 1387.7, 1374.4,
1357.3, 1433.3, 1386.5, 1575.7, 1531.8, 1560.9, 1810.4, 1657.2,
1509.4, 2110.4, 1974.5, 1856.1, 1939.6, 1832.3, 1932.5), CTACK = c(798.9,
857.5, 702.9, 707.6, 903.7, 749.9, 689.2, 734.1, 716.8, 645.2,
641.2, 663.1, 733.3, 667.3, 774.3, 782.2, 1766.7, 679.2, 1950.6,
2084.9, 2021.5, 1930.8, 1952.3, 1917.2, 723.9, 709.2, 670.7,
716.4, 944.9, 655.7, 706.8, 734.2, 680.5, 720.4, 778.3, 790.2,
962.5, 726.4, 761.9, 849.2, 701.1, 664.3), Endostatin = c(30563,
30006.5, 25972.2, 28048.9, 24503.9, 30371.2, 49733.9, 54933.8,
54293.3, 60007.4, 60403.9, 58870.4, 56801, 59752.6, 54336.3,
47746, 21270.4, 67312.9, 61931.5, 66395.3, 65680.8, 64419.4,
63415, 67230.7, 49444.2, 55122.2, 52333, 62328.7, 47513.9, 61530,
52549.8, 53633, 53983.7, 49023.8, 47305.6, 50569.8, 58754.1,
65727.9, 78382, 68290.8, 69386.8, 59982.4), `TIMP-1` = c(279.4,
262.4, 295.9, 273.3, 294.9, 278.3, 279.4, 238.7, 224.2, 239.3,
235.7, 251.7, 228.2, 237.3, 258.4, 262.4, 389.4, 237.8, 303.1,
296.4, 295.4, 297, 317, 351.6, 310.4, 244.2, 250.8, 261.1, 283.1,
251.4, 242.2, 223.3, 234.6, 291.2, 262.3, 249.7, 293.9, 259.4,
238.9, 257.9, 248.2, 329.3), tPA = c(377.9, 326.9, 346.3, 318.3,
314, 348.2, 291.2, 427.8, 433.1, 428.8, 370.9, 390, 354, 354.4,
368.8, 364.7, 595.2, 400.1, 354, 356.8, 368.7, 363.5, 394.3,
412.3, 319, 406.9, 415.4, 292.3, 320.9, 348.2, 532, 457.4, 417.3,
404, 401.5, 533.5, 213.6, 346.7, 310.7, 292.6, 280.2, 287.1),
`EG-VEGF` = c(106.8, 102.2, 107, 108.1, 138.8, 107.5, 88.1,
85.7, 82.7, 91.7, 98.8, 91.1, 82.3, 90.3, 89.2, 94.9, 128.3,
94.6, 529.1, 679.3, 599.4, 644.5, 674.1, 584.5, 98.2, 101.1,
97.3, 95.3, 86.8, 92.5, 101.9, 94.9, 109, 94.1, 103.8, 98.2,
92, 98.8, 91.6, 92.9, 98.8, 105.8), `TIMP-2` = c(466.5, 420.7,
479.8, 423, 462.7, 393.1, 400.5, 380.3, 355.9, 393.7, 423.6,
401.6, 364.2, 372.8, 391.3, 480.6, 1171.4, 357.8, 533.7,
522.6, 523.1, 511.5, 511, 644.8, 405.3, 398.2, 393, 369.7,
427.8, 380, 396.3, 394.5, 407, 490.9, 435.6, 402.7, 424.3,
436, 391.6, 405.6, 414.2, 536.2), `TGF-b1` = c(1451.2, 1194.6,
1521.3, 1705.8, 1945.4, 1900.2, 1839.4, 1701.9, 1752.9, 1714.6,
1597.6, 1660.5, 1786.9, 1694, 1519.6, 1906.1, 1654.2, 1636,
1566.8, 1730.9, 1496.2, 1850.9, 1715.1, 1648.8, 1834.9, 1686,
1769.4, 1750.8, 1833, 1862.6, 1511, 1505.9, 1524.9, 1514,
1391.2, 1369.1, 2064.6, 1790.9, 1799.1, 1548.2, 1839, 1667
), `VEGF sR3` = c(3094.3, 3235, 3116.3, 3027.1, 2536, 3061.8,
3215.7, 3189.6, 3139.1, 3618.8, 3585.6, 3556, 3326.7, 3167.5,
3403.3, 2583.1, 1621.6, 3415.2, 2293.9, 2508.4, 2662.6, 2486.2,
2538.8, 2452.5, 3012.5, 2920.1, 2914.1, 2870, 2455.2, 2809.3,
3534.8, 3501.9, 3691.4, 3517.6, 3414, 3405.6, 1943.3, 2334.2,
2137.4, 2185.7, 2273.2, 2304.9), C5 = c(5566.9, 5466.7, 5591.6,
5552.7, 5348.3, 5388.4, 5499.1, 5834.1, 5556.7, 5737.5, 5632.2,
5886.4, 5543.2, 6134.6, 5597.3, 5557.9, 5446.3, 5863.7, 5699.8,
5263.7, 5773.3, 5313.7, 6014.2, 5453.3, 6284.5, 5658.5, 5491.2,
5855.1, 5477.2, 5815.2, 5938.1, 5660.6, 6190.9, 5626.5, 6248.6,
5673, 6071.4, 6262.5, 5649.1, 6052, 6000.4, 5649), `Apo E` = c(3351.8,
936.5, 928.1, 3096.5, 2267.6, 2217.1, 2143.2, 2547.1, 2368.6,
2531.5, 1922.4, 2134.3, 2283.7, 2131, 2260.3, 2249.8, 2001.2,
2271.8, 3635.3, 2338.7, 2301.2, 2705, 2604.3, 2738.5, 5091.4,
3638.9, 2710.4, 3605, 3683, 3016.1, 3698.2, 3050.5, 2162.1,
2086.9, 2500.5, 2449.1, 4416, 2628.9, 2902.4, 3100.8, 2433.7,
3083.3), BDNF = c(325, 324, 337.9, 436.6, 557.3, 379.5, 556.2,
458.3, 400.3, 447.4, 448.1, 450.3, 425.1, 456.4, 482.1, 585,
663.2, 424.9, 343.8, 337.2, 328.1, 417.9, 343.2, 386.1, 590.6,
359.9, 340.3, 360.5, 394.1, 374.6, 367.9, 294.1, 319.4, 320,
366.2, 372.1, 502.1, 403.1, 432.9, 397.3, 389, 364.5), `bFGF-R` = c(12689.3,
13327.1, 12159.8, 11959.7, 11430.1, 11628.7, 11430.7, 12808.9,
11439, 12836.8, 13735.2, 12351.3, 11754, 12071.3, 11841.7,
10368.6, 12122.4, 12371.7, 11184.5, 11499.1, 11687.9, 10997.4,
11006.7, 10709.3, 11615, 12553.3, 11459.9, 12403, 10952,
12060.3, 13330.8, 12688.8, 13717.2, 11868.7, 11919.1, 11584.5,
10987.7, 12370.3, 11619.4, 11737.2, 12695.7, 12403.2), C8 = c(1868.7,
1678.4, 1916.9, 1966.6, 1724.9, 2028, 1840.3, 1797.4, 1871.4,
1967.9, 1926.4, 1860.2, 1860.4, 1901.4, 1951.1, 1675.4, 1026,
2024.3, 1950.6, 2174.5, 2266.5, 2251.5, 2176.4, 2067.2, 1968,
2187.1, 2360.5, 2057.5, 1971.1, 2202.2, 2099.5, 2337.2, 2320.8,
2232.9, 2358.2, 2374.8, 2071.3, 2773.3, 2695.9, 2641.2, 2738.7,
2607), `Cathepsin G` = c(1434.9, 1347, 1410.5, 1421.1, 2318.2,
1675, 1025.9, 1217.4, 1021.9, 1088.4, 1034.2, 1099.8, 1080.1,
1051.2, 967.8, 3874.4, 1875.6, 1058.9, 1093.7, 1083.2, 922.1,
1204.4, 1126.1, 1157.9, 1642.2, 1528.4, 1627.3, 1811.6, 1498.4,
1759.8, 926.3, 1022.4, 1106.7, 1058.7, 999.1, 995.3, 1020.5,
1085.4, 1088.9, 961.9, 1135.7, 1080.7), `CXCL16, soluble` = c(6878.8,
7479.3, 7088.7, 6739.4, 6143.9, 7477.5, 6742, 6822.4, 6164.9,
7320.9, 7459.1, 7150.5, 6407.1, 6504, 6901.8, 5929.6, 3924.1,
7827.3, 7358.7, 7758.3, 8305.5, 7586.9, 7829.1, 8268.3, 6860.7,
7470.7, 6857, 7003.7, 6364.7, 7738.7, 5441.6, 5609.3, 5732.5,
5128.2, 5642.7, 5402.4, 5170.2, 6272, 5963, 5996.4, 6461,
6721.3), `FGF-10` = c(255.2, 210, 262.6, 303.6, 219.4, 301.6,
241.2, 283.5, 248.6, 260.1, 265.7, 299.5, 268.9, 265.3, 201.3,
190.3, 296.6, 239.5, 242.7, 287.3, 216.2, 288.8, 283.8, 246.7,
255.1, 268.6, 236.3, 258.4, 206.8, 259.4, 264.1, 233.2, 277.1,
205.7, 200, 193.1, 251, 228, 229.9, 213.6, 247.2, 235.2),
`FGF-8B` = c(712.8, 648.4, 704.5, 763.9, 837.4, 795.9, 836.1,
841.7, 872.4, 860.1, 921.1, 845.8, 843.6, 875.2, 850, 847.4,
934.7, 938.6, 715.6, 823.9, 733.3, 801.1, 800.3, 819.9, 791.1,
930.8, 938.3, 869.6, 794.8, 939.9, 925.9, 971.2, 1015.9,
928.8, 851.9, 827, 833.6, 837.7, 858, 828.2, 935.7, 900.5
), GIIE = c(270, 254.6, 268.7, 279, 224.4, 293.2, 239.8,
238, 222.9, 260.2, 282.3, 262.3, 239, 241, 238.7, 200.6,
207.3, 252, 270.8, 281.1, 354.2, 289.2, 293, 295.2, 274.5,
291.4, 270.6, 275.8, 232, 272.7, 267, 245.3, 278.3, 260.6,
264, 250.6, 232.9, 274.4, 256.2, 254.1, 271.2, 282), GV = c(433.7,
441.1, 438.9, 468, 425.6, 459.2, 317.6, 332.2, 326, 306.4,
307.4, 310.6, 347.9, 317, 273.1, 325.9, 798.4, 299.1, 327.8,
307.9, 258.6, 308.7, 306.6, 298.6, 319.2, 326.2, 299.8, 329,
436.5, 297.7, 320.9, 306.6, 314.1, 312.1, 298.1, 300.1, 417,
306.6, 314.7, 321.4, 304.8, 305), `IL-12` = c(725.7, 667,
734.8, 772.8, 1045.1, 829.4, 659.4, 695.3, 653.8, 672.8,
701.1, 658.1, 683.5, 670.8, 678.8, 1002.5, 991.4, 703.8,
667.9, 714.1, 630.6, 720, 689.8, 781.4, 671.2, 715.2, 748.9,
693.2, 723.5, 724.7, 868.6, 891.2, 917.9, 858.8, 868.9, 828.6,
744.7, 711.9, 715.1, 683.2, 740.2, 724.3)), row.names = c(NA,
-42L), class = c("tbl_df", "tbl", "data.frame"))
我的原始数据包含大量变量 (1,130)，其中我使用 library(dplyr) 汇总函数按时间点为每一列创建平均值。
df_mean <- df %>%
group_by(TimePoint) %>%
summarise_at(vars(C4b:GV), mean, na.rm = T)
现在我想创建一个新的数据框，它只是 df 中的原始值与所有变量的每个时间点的平均值之间的差异，因此本质上是 df - df_mean = 新数据框，变量“day”除外，“主题”和“时间点”。
谢谢！
答案 0 :(得分：1)
library(tidyverse)
df_centered <- df %>%
group_by(TimePoint) %>%
mutate_at(vars(!c("day", "Subject")), ~scale(., scale = FALSE), na.rm = TRUE)
答案 1 :(得分：1)
我试过这样的东西，希望它有帮助。
df1 <- df[,-c(1,2,25)]
df_mm <- full_join(df1, df_mean, by = "TimePoint")
df_ss <- NULL
names <- colnames(df)[-length(df)]
for(i in 2:length(df1)){
df_ss[[i]] <- (df_mm[,i] - df_mm[,length(df1)+i-1])
}
df_ss[1] <- NULL
df_ss <- do.call(cbind, df_ss)
df_ss <- mutate(df_ss,
day = df$day,
Subject = df$Subject,
TimePoint = df$TimePoint)
df_ss <- df_ss[,c(22:24,1:21)]
colnames(df_ss) <- names
View(df_ss)