I am trying to correlate the columns of TRE, Run-in, and Control together. However, I want to do it by the values of another column (so it would only be correlating certain rows at a time)
Here is my data:
change<-structure(list(Subject.ID = c(168236L, 168236L, 168236L, 168236L,
168236L, 168236L, 168236L, 168236L, 168236L, 168695L, 168695L,
168695L, 168695L, 168695L, 168695L, 168695L, 168695L, 168695L,
168695L, 168696L, 168696L, 168696L, 168696L, 168696L, 168696L,
168696L, 168696L, 168696L, 168696L, 168914L, 168914L, 168914L,
168914L, 168914L, 168914L, 168914L, 168914L, 168914L, 169242L,
169242L, 169242L, 169242L, 169242L, 169242L, 169242L, 169242L,
169242L, 169242L, 169728L, 169728L, 169728L, 169728L, 169728L,
169728L, 169728L, 169728L, 169728L, 169728L, 170992L, 170992L,
170992L, 170992L, 170992L, 170992L, 170992L, 170992L, 170992L,
170992L, 172482L, 172482L, 172482L, 172482L, 172482L, 172482L,
172482L, 172482L, 172482L, 172482L, 172483L, 172483L, 172483L,
172483L, 172483L, 172483L, 172483L, 172483L, 172483L, 172483L,
172490L, 172490L, 172490L, 172490L, 172490L, 172490L, 172490L,
172490L, 172490L, 172490L), Result.Test.Mnemonic = c("Glucose",
"Chol", "Trig", "HDL", "LDL Direct", "NONHDLCHOL", "VLDL C",
"High Sens CRP", "Insulin Fasting", "Glucose", "Chol", "Trig",
"HDL", "LDL Direct", "NONHDLCHOL", "VLDL C", "High Sens CRP",
"Insulin Fasting", "C Peptide", "Glucose", "Chol", "Trig", "HDL",
"LDL Direct", "VLDL C", "NONHDLCHOL", "High Sens CRP", "Insulin Fasting",
"C Peptide", "Glucose", "Chol", "Trig", "HDL", "LDL Direct",
"NONHDLCHOL", "VLDL C", "High Sens CRP", "Insulin Fasting", "Glucose",
"Chol", "Trig", "HDL", "LDL Direct", "NONHDLCHOL", "VLDL C",
"High Sens CRP", "Insulin Fasting", "C Peptide", "Glucose", "Chol",
"Trig", "HDL", "LDL Direct", "NONHDLCHOL", "VLDL C", "High Sens CRP",
"Insulin Fasting", "C Peptide", "Glucose", "Chol", "Trig", "HDL",
"LDL Direct", "VLDL C", "NONHDLCHOL", "High Sens CRP", "Insulin Fasting",
"C Peptide", "Glucose", "Chol", "Trig", "HDL", "LDL Direct",
"NONHDLCHOL", "VLDL C", "High Sens CRP", "Insulin Fasting", "C Peptide",
"Glucose", "Chol", "Trig", "HDL", "LDL Direct", "VLDL C", "NONHDLCHOL",
"High Sens CRP", "Insulin Fasting", "C Peptide", "Glucose", "Chol",
"Trig", "HDL", "LDL Direct", "NONHDLCHOL", "VLDL C", "High Sens CRP",
"Insulin Fasting", "C Peptide"), `Run-in` = c(83, 137, 68, 50,
79, 87, 8, 1.3, 10, 96, 149, 96, 57, 78, 92, 14, 4.6, 12, 2.2,
99, 203, 68, 65, 129, 9, 138, 6.3, 10, 1.5, NA, NA, NA, NA, NA,
NA, NA, NA, NA, 84, 141, 119, 38, 87, 103, 16, 0.9, 15, 2.6,
90, 163, 77, 48, 103, 115, 12, NA, 9, 1.9, 102, 177, 126, 38,
118, 21, 139, 2.9, 18, 3, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, 122, 207, 144, 36, 152, 19, 171, 4, 55, 4.4, 100, 186, 247,
39, 111, 147, 36, 3.5, 24, 3.1), TRE = c(NA, NA, NA, NA, NA,
NA, NA, NA, NA, 91, 149, 80, 57, 80, 92, 12, 5.4, 7, 1.7, 98,
181, 80, 52, 117, 12, 129, 1, 14, 1.7, 95, 274, 228, 46, 197,
228, 31, 3, 24, 88, 123, 82, 39, 77, 84, 7, 0.7, 16, 2.3, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, 101, 172, 83, 44, 117, 11,
128, 11.3, 14, 1.9, 87, 175, 127, 46, 109, 129, 20, 1.7, 10,
2.1, 111, 201, 94, 37, 150, 14, 164, 3.6, 37, 3.3, 98, 200, 344,
39, 109, 161, 52, 3.2, 21, 2.6), Control = c(NA, NA, NA, NA,
NA, NA, NA, NA, NA, 96, 140, 59, 55, 75, 85, 10, 3.2, 8, 1.8,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, 93, 137, 79, 39, 90, 98, 8, 1, 24, 2.5, 93, 152,
60, 49, 94, 103, 9, NA, 6, 1.5, 106, 167, 90, 42, 112, 13, 125,
17, 13, 2, 86, 165, 112, 52, 102, 113, 11, 1.6, 19, 2.2, 107,
188, 104, 34, 137, 17, 154, 4.6, 36, 3.6, 104, 188, 403, 37,
97, 151, 54, 6, 35, 3.4), Change_Run_Control = c(NA, NA, NA,
NA, NA, NA, NA, NA, NA, 0, 9, 37, 2, 3, 7, 4, 1.4, 4, 0.4, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, 9, 4, 40, 1, 3, 5, 8, 0.1, 9, 0.1, 3, 11, 17, 1, 9, 12,
3, NA, 3, 0.4, 4, 10, 36, 4, 6, 8, 14, 14.1, 5, 1, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, 15, 19, 40, 2, 15, 2, 17, 0.6, 19,
0.8, 4, 2, 156, 2, 14, 4, 18, 2.5, 11, 0.3), Change_Run_TRE = c(NA,
NA, NA, NA, NA, NA, NA, NA, NA, 5, 0, 16, 0, 2, 0, 2, 0.800000000000001,
5, 0.5, 1, 22, 12, 13, 12, 3, 9, 5.3, 4, 0.2, NA, NA, NA, NA,
NA, NA, NA, NA, NA, 4, 18, 37, 1, 10, 19, 9, 0.2, 1, 0.3, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, 1, 5, 43, 6, 1, 10, 11, 8.4,
4, 1.1, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 11, 6, 50, 1,
2, 5, 7, 0.4, 18, 1.1, 2, 14, 97, 0, 2, 14, 16, 0.3, 3, 0.5),
Change_Control_TRE = c(NA, NA, NA, NA, NA, NA, NA, NA, NA,
5, 9, 21, 2, 5, 7, 2, 2.2, 1, 0.1, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, 5, 14,
3, 0, 13, 14, 1, 0.3, 8, 0.2, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, 5, 5, 7, 2, 5, 2, 3, 5.7, 1, 0.1, 1, 10, 15,
6, 7, 16, 9, 0.0999999999999999, 9, 0.1, 4, 13, 10, 3, 13,
3, 10, 1, 1, 0.3, 6, 12, 59, 2, 12, 10, 2, 2.8, 14, 0.8)), row.names = c(NA,
-98L), class = c("tbl_df", "tbl", "data.frame"))
Here is what one of the correlations looks like outside the loop (and it functions):
cor.test(change$`Run-in`[change$Result.Test.Mnemonic== "C Peptide"] , change$Control[change$Result.Test.Mnemonic== "C Peptide"])
I can't get my loop to work. Here is what I have so far:
# Define phases and variables
phases <- c("`Run-in`", "Control", "TRE")
variables <- c(unique(change$Result.Test.Mnemonic))
# Loop over phases and variables
for (variable in variables) {
for (i in 1:(length(phases)-1)) {
for (j in (i+1):length(phases)) {
(paste0("correlation<-cor.test(",
"change$", phases[i], "[change$Result.Test.Mnemonic == '", variable, "'],change$", phases[j],
"[change$Result.Test.Mnemonic== '", variable, "'])"))
}
}
}
The syntax seems to run, but it doesn't output anything that I can visually see to assess the correlations. I think it has something to do with how I am using the paste0 function. But I am not sure what I am doing wrong.
I tried using the paste0 function inside of the cor.test function (using it twice for each side of the correlation). But it did not work. It seemed to throw an error because cor.test needs numeric data given to it.
Try something like this which stores the results in a list() and evaluates each string as an expression:
phases <- c("`Run-in`", "Control", "TRE")
variables <- c(unique(change$Result.Test.Mnemonic))
list_results <- list()
# Loop over phases and variables
for (variable in variables) {
for (i in 1:(length(phases)-1)) {
for (j in (i+1):length(phases)) {
x <- (eval(str2lang(paste0("correlation<-cor.test(",
"change$", phases[i], "[change$Result.Test.Mnemonic == '", variable, "'],change$", phases[j],
"[change$Result.Test.Mnemonic== '", variable, "'])"))))
list_results <- append(list_results, list(x))
}
}
}
list_results
#> [[1]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Glucose"] and change$Control[change$Result.Test.Mnemonic == "Glucose"]
#> t = 3.193, df = 4, p-value = 0.03312
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.1150259 0.9829707
#> sample estimates:
#> cor
#> 0.8474749
#>
#>
#> [[2]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Glucose"] and change$TRE[change$Result.Test.Mnemonic == "Glucose"]
#> t = 7.2959, df = 4, p-value = 0.001876
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.703402 0.996239
#> sample estimates:
#> cor
#> 0.9644203
#>
#>
#> [[3]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "Glucose"] and change$TRE[change$Result.Test.Mnemonic == "Glucose"]
#> t = 3.9877, df = 4, p-value = 0.0163
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.2997933 0.9884098
#> sample estimates:
#> cor
#> 0.8938744
#>
#>
#> [[4]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Chol"] and change$Control[change$Result.Test.Mnemonic == "Chol"]
#> t = 6.6847, df = 4, p-value = 0.002604
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.6583485 0.9955516
#> sample estimates:
#> cor
#> 0.958039
#>
#>
#> [[5]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Chol"] and change$TRE[change$Result.Test.Mnemonic == "Chol"]
#> t = 4.2587, df = 4, p-value = 0.01307
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.3526332 0.9896963
#> sample estimates:
#> cor
#> 0.9051543
#>
#>
#> [[6]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "Chol"] and change$TRE[change$Result.Test.Mnemonic == "Chol"]
#> t = 8.0578, df = 4, p-value = 0.001288
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.7487535 0.9968957
#> sample estimates:
#> cor
#> 0.9705506
#>
#>
#> [[7]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Trig"] and change$Control[change$Result.Test.Mnemonic == "Trig"]
#> t = 6.9099, df = 4, p-value = 0.002301
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.6759846 0.9958250
#> sample estimates:
#> cor
#> 0.9605728
#>
#>
#> [[8]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Trig"] and change$TRE[change$Result.Test.Mnemonic == "Trig"]
#> t = 4.6785, df = 4, p-value = 0.009458
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.4253719 0.9913137
#> sample estimates:
#> cor
#> 0.9195057
#>
#>
#> [[9]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "Trig"] and change$TRE[change$Result.Test.Mnemonic == "Trig"]
#> t = 18.708, df = 4, p-value = 4.806e-05
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.9468272 0.9994091
#> sample estimates:
#> cor
#> 0.994334
#>
#>
#> [[10]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "HDL"] and change$Control[change$Result.Test.Mnemonic == "HDL"]
#> t = 6.3865, df = 4, p-value = 0.003085
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.6329137 0.9951470
#> sample estimates:
#> cor
#> 0.9543003
#>
#>
#> [[11]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "HDL"] and change$TRE[change$Result.Test.Mnemonic == "HDL"]
#> t = 3.8004, df = 4, p-value = 0.0191
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.2603709 0.9873808
#> sample estimates:
#> cor
#> 0.8849385
#>
#>
#> [[12]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "HDL"] and change$TRE[change$Result.Test.Mnemonic == "HDL"]
#> t = 4.7233, df = 4, p-value = 0.00915
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.4325193 0.9914639
#> sample estimates:
#> cor
#> 0.9208482
#>
#>
#> [[13]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "LDL Direct"] and change$Control[change$Result.Test.Mnemonic == "LDL Direct"]
#> t = 9.5007, df = 4, p-value = 0.000685
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.8112529 0.9977474
#> sample estimates:
#> cor
#> 0.978553
#>
#>
#> [[14]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "LDL Direct"] and change$TRE[change$Result.Test.Mnemonic == "LDL Direct"]
#> t = 9.6744, df = 4, p-value = 0.0006388
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.8172278 0.9978258
#> sample estimates:
#> cor
#> 0.9792925
#>
#>
#> [[15]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "LDL Direct"] and change$TRE[change$Result.Test.Mnemonic == "LDL Direct"]
#> t = 6.1285, df = 4, p-value = 0.003592
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.6087787 0.9947513
#> sample estimates:
#> cor
#> 0.9506571
#>
#>
#> [[16]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "NONHDLCHOL"] and change$Control[change$Result.Test.Mnemonic == "NONHDLCHOL"]
#> t = 7.6117, df = 4, p-value = 0.001599
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.7234910 0.9965341
#> sample estimates:
#> cor
#> 0.9671709
#>
#>
#> [[17]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "NONHDLCHOL"] and change$TRE[change$Result.Test.Mnemonic == "NONHDLCHOL"]
#> t = 5.725, df = 4, p-value = 0.004608
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.5665669 0.9940305
#> sample estimates:
#> cor
#> 0.9440503
#>
#>
#> [[18]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "NONHDLCHOL"] and change$TRE[change$Result.Test.Mnemonic == "NONHDLCHOL"]
#> t = 6.7526, df = 4, p-value = 0.002508
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.6638006 0.9956367
#> sample estimates:
#> cor
#> 0.9588273
#>
#>
#> [[19]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "VLDL C"] and change$Control[change$Result.Test.Mnemonic == "VLDL C"]
#> t = 6.7504, df = 4, p-value = 0.002511
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.6636268 0.9956340
#> sample estimates:
#> cor
#> 0.9588023
#>
#>
#> [[20]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "VLDL C"] and change$TRE[change$Result.Test.Mnemonic == "VLDL C"]
#> t = 3.7879, df = 4, p-value = 0.0193
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.2576597 0.9873077
#> sample estimates:
#> cor
#> 0.8843066
#>
#>
#> [[21]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "VLDL C"] and change$TRE[change$Result.Test.Mnemonic == "VLDL C"]
#> t = 7.5296, df = 4, p-value = 0.001666
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.7184565 0.9964608
#> sample estimates:
#> cor
#> 0.9664869
#>
#>
#> [[22]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "High Sens CRP"] and change$Control[change$Result.Test.Mnemonic == "High Sens CRP"]
#> t = 0.1571, df = 3, p-value = 0.8851
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> -0.8605146 0.9008071
#> sample estimates:
#> cor
#> 0.09033273
#>
#>
#> [[23]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "High Sens CRP"] and change$TRE[change$Result.Test.Mnemonic == "High Sens CRP"]
#> t = -0.17785, df = 4, p-value = 0.8675
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> -0.8397710 0.7789816
#> sample estimates:
#> cor
#> -0.08857772
#>
#>
#> [[24]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "High Sens CRP"] and change$TRE[change$Result.Test.Mnemonic == "High Sens CRP"]
#> t = 5.2691, df = 4, p-value = 0.006216
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.5112962 0.9930266
#> sample estimates:
#> cor
#> 0.9349158
#>
#>
#> [[25]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Insulin Fasting"] and change$Control[change$Result.Test.Mnemonic == "Insulin Fasting"]
#> t = 2.3747, df = 4, p-value = 0.07643
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> -0.1231010 0.9726627
#> sample estimates:
#> cor
#> 0.7648743
#>
#>
#> [[26]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "Insulin Fasting"] and change$TRE[change$Result.Test.Mnemonic == "Insulin Fasting"]
#> t = 6.449, df = 4, p-value = 0.002976
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.6384538 0.9952362
#> sample estimates:
#> cor
#> 0.9551233
#>
#>
#> [[27]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "Insulin Fasting"] and change$TRE[change$Result.Test.Mnemonic == "Insulin Fasting"]
#> t = 3.1432, df = 4, p-value = 0.03474
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.1018870 0.9825158
#> sample estimates:
#> cor
#> 0.843686
#>
#>
#> [[28]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "C Peptide"] and change$Control[change$Result.Test.Mnemonic == "C Peptide"]
#> t = 3.3098, df = 4, p-value = 0.02966
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.1450989 0.9839733
#> sample estimates:
#> cor
#> 0.8558752
#>
#>
#> [[29]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$`Run-in`[change$Result.Test.Mnemonic == "C Peptide"] and change$TRE[change$Result.Test.Mnemonic == "C Peptide"]
#> t = 4.1196, df = 4, p-value = 0.01462
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.3261270 0.9890636
#> sample estimates:
#> cor
#> 0.8995915
#>
#>
#> [[30]]
#>
#> Pearson's product-moment correlation
#>
#> data: change$Control[change$Result.Test.Mnemonic == "C Peptide"] and change$TRE[change$Result.Test.Mnemonic == "C Peptide"]
#> t = 6.0162, df = 4, p-value = 0.003844
#> alternative hypothesis: true correlation is not equal to 0
#> 95 percent confidence interval:
#> 0.5976075 0.9945642
#> sample estimates:
#> cor
#> 0.9489383
Created on 2024-03-20 with reprex v2.1.0