| title | author | date | output | editor_options | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
NONMEM workshop |
Sungpil Han, Kyun-Seop Bae |
2019-02-07 |
|
|
https://github.com/asancpt/edison-nmw
license: GPL-3
NONMEM Workshop 2017, 2018에서 사용된 nmw 패키지를 사용한 Edison 사이언스 앱입니다.
A table (head) and a figure of input dataset is shown below.
kable(inputFirst, format = "markdown")| name | value |
|---|---|
| Dataset | Theoph |
| Method | COND |
| nTheta | 3 |
| nEta | 3 |
| nEps | 2 |
| THETAinit | 2, 50, 0.1 |
| OMinit | 0.2, 0.1, 0.1, 0.1, 0.2, 0.1, 0.1, 0.1, 0.2 |
| SGinit | 0.1, 0, 0, 0.1 |
kable(head(DataAll, n = 20), caption = "input data", format = "markdown")| ID | TIME | DV |
|---|---|---|
| 1 | 0.00 | 0.74 |
| 1 | 0.25 | 2.84 |
| 1 | 0.57 | 6.57 |
| 1 | 1.12 | 10.50 |
| 1 | 2.02 | 9.66 |
| 1 | 3.82 | 8.58 |
| 1 | 5.10 | 8.36 |
| 1 | 7.03 | 7.47 |
| 1 | 9.05 | 6.89 |
| 1 | 12.12 | 5.94 |
| 1 | 24.37 | 3.28 |
| 2 | 0.00 | 0.00 |
| 2 | 0.27 | 1.72 |
| 2 | 0.52 | 7.91 |
| 2 | 1.00 | 8.31 |
| 2 | 1.92 | 8.33 |
| 2 | 3.50 | 6.85 |
| 2 | 5.02 | 6.08 |
| 2 | 7.03 | 5.40 |
| 2 | 9.00 | 4.55 |
if (NMDataset == "Emax") xyplot(DV ~ log(CE) | ID, data=DataAll, type="b")
if (NMDataset == "Theoph") {
xyplot(DV~TIME|ID,
data=DataAll,
type="b",
index.cond=list(order(as.numeric(levels(unique(DataAll$ID))))))
}
- Dataset: Theoph
- Method: COND
PREDFILE <- ifelse(NMDataset == "Emax", "03-Emax/PRED.R", "04-THEO/PRED.R")
source(PREDFILE)
FGD <- deriv(expr = ~ DOSE/
(TH2*exp(ETA2)) *
TH1*exp(ETA1)/
(TH1*exp(ETA1) - TH3*exp(ETA3)) *
(exp(-TH3*exp(ETA3)*TIME)-exp(-TH1*exp(ETA1)*TIME)),
namevec = c("ETA1","ETA2","ETA3"),
function.arg = c("TH1", "TH2", "TH3",
"ETA1", "ETA2", "ETA3",
"DOSE", "TIME"),
func=TRUE,
hessian=TRUE)
H <- deriv(~F + F*EPS1 + EPS2, c("EPS1", "EPS2"),
function.arg=c("F", "EPS1", "EPS2"),
func=TRUE)
InitPara <- nmw::InitStep(DataAll,
THETAinit=THETAinit,
OMinit=OMinit,
SGinit=SGinit,
LB=LB,
UB=UB,
Pred=PRED,
METHOD=METHOD)
(EstRes = EstStep()) # 0.6200359 secs, 0.4930282 secs## $`Initial OFV`
## [1] 188.7005
##
## $Time
## Time difference of 2.053784 mins
##
## $Optim
## $Optim$par
## [1] -0.191038342 -0.331507105 -0.036604792 0.490015633 0.038777478
## [6] -0.004544747 -1.148169722 0.087407760 -1.321560134 -0.772014649
## [11] -0.019855262
##
## $Optim$value
## [1] 92.2179
##
## $Optim$counts
## function gradient
## 64 64
##
## $Optim$convergence
## [1] 0
##
## $Optim$message
## [1] "CONVERGENCE: REL_REDUCTION_OF_F <= FACTR*EPSMCH"
##
##
## $`Final Estimates`
## [1] 1.494974789 32.477041485 0.087231491 0.436308094 0.057274486
## [6] 0.019876368 -0.006712609 0.011663141 0.020602621 0.017481460
## [11] 0.078685560
(CovRes = CovStep())## $Time
## Time difference of 54.10309 secs
##
## $`Standard Error`
## [1] 0.343329779 1.707724512 0.003845457 0.240759851 0.048796392
## [6] 0.023296233 0.042579379 0.018214087 0.034838779 0.009644326
## [11] 0.062764272
##
## $`Covariance Matrix of Estimates`
## [,1] [,2] [,3] [,4]
## [1,] 0.1178753370 0.3257709609 -3.138296e-04 0.0449271375
## [2,] 0.3257709609 2.9163230078 9.528463e-04 0.0117646115
## [3,] -0.0003138296 0.0009528463 1.478754e-05 -0.0003818976
## [4,] 0.0449271375 0.0117646115 -3.818976e-04 0.0579653056
## [5,] 0.0046498909 -0.0289574329 -6.588987e-05 0.0085920048
## [6,] -0.0015850872 -0.0082040885 -8.255648e-06 0.0001760737
## [7,] -0.0013216131 0.0219967109 9.812688e-05 -0.0050430722
## [8,] 0.0009423561 0.0047046258 -3.935965e-06 0.0008179267
## [9,] -0.0004880641 -0.0190762111 -8.911725e-05 0.0025871642
## [10,] -0.0003309453 0.0024072682 1.802441e-05 -0.0014994314
## [11,] 0.0005763975 -0.0403410196 -1.528631e-04 0.0089227863
## [,5] [,6] [,7] [,8]
## [1,] 4.649891e-03 -1.585087e-03 -1.321613e-03 9.423561e-04
## [2,] -2.895743e-02 -8.204088e-03 2.199671e-02 4.704626e-03
## [3,] -6.588987e-05 -8.255648e-06 9.812688e-05 -3.935965e-06
## [4,] 8.592005e-03 1.760737e-04 -5.043072e-03 8.179267e-04
## [5,] 2.381088e-03 6.754867e-04 -4.131683e-04 -3.910530e-04
## [6,] 6.754867e-04 5.427145e-04 3.708019e-04 -3.819689e-04
## [7,] -4.131683e-04 3.708019e-04 1.813004e-03 -4.546660e-04
## [8,] -3.910530e-04 -3.819689e-04 -4.546660e-04 3.317530e-04
## [9,] 9.992286e-04 4.790779e-04 -6.173928e-04 -2.762646e-04
## [10,] -3.528927e-04 -1.223755e-04 1.614071e-04 5.439237e-05
## [11,] 1.934613e-03 4.362098e-04 -1.919342e-03 -3.893778e-05
## [,9] [,10] [,11]
## [1,] -4.880641e-04 -3.309453e-04 5.763975e-04
## [2,] -1.907621e-02 2.407268e-03 -4.034102e-02
## [3,] -8.911725e-05 1.802441e-05 -1.528631e-04
## [4,] 2.587164e-03 -1.499431e-03 8.922786e-03
## [5,] 9.992286e-04 -3.528927e-04 1.934613e-03
## [6,] 4.790779e-04 -1.223755e-04 4.362098e-04
## [7,] -6.173928e-04 1.614071e-04 -1.919342e-03
## [8,] -2.762646e-04 5.439237e-05 -3.893778e-05
## [9,] 1.213741e-03 -2.468340e-04 1.842491e-03
## [10,] -2.468340e-04 9.301303e-05 -4.670010e-04
## [11,] 1.842491e-03 -4.670010e-04 3.939354e-03
##
## $`Correlation Matrix of Estimates`
## [,1] [,2] [,3] [,4] [,5]
## [1,] 1.00000000 0.55562667 -0.23770290 0.54351710 0.2775515
## [2,] 0.55562667 1.00000000 0.14509654 0.02861381 -0.3474998
## [3,] -0.23770290 0.14509654 1.00000000 -0.41249143 -0.3511422
## [4,] 0.54351710 0.02861381 -0.41249143 1.00000000 0.7313457
## [5,] 0.27755150 -0.34749978 -0.35114216 0.73134574 1.0000000
## [6,] -0.19817825 -0.20621812 -0.09215470 0.03139242 0.5942147
## [7,] -0.09040523 0.30251061 0.59929505 -0.49193962 -0.1988566
## [8,] 0.15069408 0.15125156 -0.05619476 0.18651893 -0.4399876
## [9,] -0.04080397 -0.32063534 -0.66519790 0.30844448 0.5877792
## [10,] -0.09994770 0.14616210 0.48600557 -0.64575926 -0.7498650
## [11,] 0.02674842 -0.37637135 -0.63334764 0.59047828 0.6316753
## [,6] [,7] [,8] [,9] [,10]
## [1,] -0.19817825 -0.09040523 0.15069408 -0.04080397 -0.0999477
## [2,] -0.20621812 0.30251061 0.15125156 -0.32063534 0.1461621
## [3,] -0.09215470 0.59929505 -0.05619476 -0.66519790 0.4860056
## [4,] 0.03139242 -0.49193962 0.18651893 0.30844448 -0.6457593
## [5,] 0.59421474 -0.19885656 -0.43998764 0.58777920 -0.7498650
## [6,] 1.00000000 0.37381514 -0.90019155 0.59027930 -0.5446745
## [7,] 0.37381514 1.00000000 -0.58625389 -0.41619732 0.3930534
## [8,] -0.90019155 -0.58625389 1.00000000 -0.43536634 0.3096412
## [9,] 0.59027930 -0.41619732 -0.43536634 1.00000000 -0.7346325
## [10,] -0.54467452 0.39305337 0.30964117 -0.73463251 1.0000000
## [11,] 0.29833020 -0.71819203 -0.03406052 0.84261639 -0.7714956
## [,11]
## [1,] 0.02674842
## [2,] -0.37637135
## [3,] -0.63334764
## [4,] 0.59047828
## [5,] 0.63167529
## [6,] 0.29833020
## [7,] -0.71819203
## [8,] -0.03406052
## [9,] 0.84261639
## [10,] -0.77149557
## [11,] 1.00000000
##
## $`Inverse Covariance Matrix of Estimates`
## [,1] [,2] [,3] [,4] [,5]
## [1,] 94.14289 -18.830548 1899.3717 245.13146 -2089.2012
## [2,] -18.83055 4.938307 -398.6819 -62.25571 487.9919
## [3,] 1899.37167 -398.681855 249350.3935 5572.98931 -38648.4406
## [4,] 245.13146 -62.255707 5572.9893 2508.34554 -16016.3036
## [5,] -2089.20123 487.991905 -38648.4406 -16016.30362 108118.3489
## [6,] 2205.58120 -444.438161 44370.0303 15507.64141 -105052.6212
## [7,] -1246.19795 228.835611 -25711.0044 -15322.08095 97740.3736
## [8,] -3282.99320 696.022606 -31002.1363 -42102.30301 269766.8179
## [9,] -1110.12057 193.788688 16402.0735 -12555.03598 83750.8997
## [10,] 946.94833 -155.992459 18774.0536 23046.33017 -135931.4313
## [11,] 85.49135 -2.328671 -11520.0492 732.60656 -6381.4473
## [,6] [,7] [,8] [,9] [,10]
## [1,] 2205.5812 -1246.1980 -3282.9932 -1110.1206 946.9483
## [2,] -444.4382 228.8356 696.0226 193.7887 -155.9925
## [3,] 44370.0303 -25711.0044 -31002.1363 16402.0735 18774.0536
## [4,] 15507.6414 -15322.0809 -42102.3030 -12555.0360 23046.3302
## [5,] -105052.6212 97740.3736 269766.8179 83750.8997 -135931.4313
## [6,] 128285.3786 -104612.4173 -252309.1008 -84958.2522 151141.7351
## [7,] -104612.4173 114145.6487 293594.5831 93543.9045 -166749.9752
## [8,] -252309.1008 293594.5831 819835.8823 259111.9627 -433959.0439
## [9,] -84958.2522 93543.9045 259111.9627 93296.2109 -135245.1663
## [10,] 151141.7351 -166749.9752 -433959.0439 -135245.1663 310124.4660
## [11,] 3298.6060 -5186.4822 -24261.0881 -12032.5091 11298.2048
## [,11]
## [1,] 85.491350
## [2,] -2.328671
## [3,] -11520.049158
## [4,] 732.606563
## [5,] -6381.447289
## [6,] 3298.606010
## [7,] -5186.482194
## [8,] -24261.088086
## [9,] -12032.509110
## [10,] 11298.204758
## [11,] 5080.122399
##
## $`Eigen Values`
## [1] 0.0009904174 0.0157407843 0.0315747784 0.0500147198 0.1104912531
## [6] 0.2311610462 0.4827072872 0.8339135712 1.7444213270 2.6912515128
## [11] 4.8077333025
##
## $`R Matrix`
## [,1] [,2] [,3] [,4] [,5]
## [1,] 18.8012513 -2.2046769 225.77985 -0.1465552 -12.697216
## [2,] -2.2046769 0.6645096 -33.05175 -0.4589011 5.717464
## [3,] 225.7798537 -33.0517452 43162.57613 149.0174270 -1420.263179
## [4,] -0.1465552 -0.4589011 149.01743 66.8376280 -341.808878
## [5,] -12.6972163 5.7174640 -1420.26318 -341.8088779 3033.820347
## [6,] 30.0931922 -10.4281894 2285.54968 370.5125005 -4764.965849
## [7,] 10.1636059 -2.6040938 655.35407 107.5304852 -400.048865
## [8,] -10.5377078 4.7219807 -552.46780 -201.8886249 1799.598586
## [9,] -4.0262676 1.9650324 -842.54472 59.9517718 -358.198859
## [10,] 110.6979157 -27.4491261 3028.27516 173.2219875 -1606.444117
## [11,] -4.0238297 -0.6323363 26.22163 13.9710247 -339.905473
## [,6] [,7] [,8] [,9] [,10]
## [1,] 30.09319 10.163606 -10.537708 -4.026268 110.69792
## [2,] -10.42819 -2.604094 4.721981 1.965032 -27.44913
## [3,] 2285.54968 655.354071 -552.467797 -842.544724 3028.27516
## [4,] 370.51250 107.530485 -201.888625 59.951772 173.22199
## [5,] -4764.96585 -400.048865 1799.598586 -358.198859 -1606.44412
## [6,] 15847.46070 541.777435 -2001.858260 625.019739 5561.25552
## [7,] 541.77743 -183.299877 -2671.889440 1301.517689 1290.60030
## [8,] -2001.85826 -2671.889440 18626.548357 -3025.249423 -6379.98554
## [9,] 625.01974 1301.517689 -3025.249423 5320.516803 2955.83753
## [10,] 5561.25552 1290.600300 -6379.985542 2955.837533 87787.85042
## [11,] 423.54214 -539.434745 -1379.136159 32.944973 6295.20060
## [,11]
## [1,] -4.0238297
## [2,] -0.6323363
## [3,] 26.2216279
## [4,] 13.9710247
## [5,] -339.9054732
## [6,] 423.5421370
## [7,] -539.4347448
## [8,] -1379.1361586
## [9,] 32.9449730
## [10,] 6295.2005997
## [11,] 1141.7530516
##
## $`S Matrix`
## [,1] [,2] [,3] [,4] [,5]
## [1,] 21.878934 -2.1617003 229.27742 -7.199371 170.56370
## [2,] -2.161700 0.6399393 -26.46949 2.736611 -19.54625
## [3,] 229.277419 -26.4694915 41684.25900 207.472758 -1573.06979
## [4,] -7.199371 2.7366114 207.47276 67.475499 -576.10963
## [5,] 170.563699 -19.5462485 -1573.06979 -576.109625 7666.82628
## [6,] -417.804918 9.2903759 5438.49660 1676.692630 -24743.03593
## [7,] 4.765143 -8.1628323 -1021.42781 270.481581 -3868.83377
## [8,] 113.994065 58.9434284 -3479.73701 -2109.525185 31171.16279
## [9,] -125.378884 -13.2858239 -16105.78614 765.131964 -12057.77607
## [10,] 341.361817 -159.6079690 36661.48584 -753.609076 -5019.10613
## [11,] -39.566288 -10.4130449 -6012.33259 -136.550596 97.67359
## [,6] [,7] [,8] [,9] [,10]
## [1,] -4.178049e+02 4.765143 113.99406 -125.37888 341.3618
## [2,] 9.290376e+00 -8.162832 58.94343 -13.28582 -159.6080
## [3,] 5.438497e+03 -1021.427808 -3479.73701 -16105.78614 36661.4858
## [4,] 1.676693e+03 270.481581 -2109.52518 765.13196 -753.6091
## [5,] -2.474304e+04 -3868.833771 31171.16279 -12057.77607 -5019.1061
## [6,] 8.827955e+04 15016.112562 -119026.38370 45682.84796 18856.8584
## [7,] 1.501611e+04 3568.984452 -25353.27346 10845.00780 1460.2168
## [8,] -1.190264e+05 -25353.273459 192088.63498 -76483.63877 -25058.1180
## [9,] 4.568285e+04 10845.007796 -76483.63877 39750.38548 -1104.7602
## [10,] 1.885686e+04 1460.216760 -25058.11800 -1104.76021 269888.0254
## [11,] 2.948473e+02 239.380485 -1640.05163 4387.70032 10736.5982
## [,11]
## [1,] -39.56629
## [2,] -10.41304
## [3,] -6012.33259
## [4,] -136.55060
## [5,] 97.67359
## [6,] 294.84733
## [7,] 239.38049
## [8,] -1640.05163
## [9,] 4387.70032
## [10,] 10736.59821
## [11,] 2910.02752
#PostHocEta() # FinalPara from EstStep()
#get("EBE", envir=e)PRED = function(THETA, ETAi, DATAi)
{
CE = DATAi[,"CE"]
CE50 = as.numeric(THETA[1] * exp(ETAi[1]))
BASE = as.numeric(THETA[2])
F = BASE * (1 - CE/(CE50 + CE))
G1 = BASE * CE * CE50 / (CE50 + CE)^2
H1 = rep(1, nrow(DATAi))
D11 = (-2*CE50/(CE + CE50) + 1) * G1
return(cbind(F, G1, H1, D11))
}
nTheta = 2
nEta = 1
nEps = 1
THETAinit = 10, 100
OMinit = 0.2
SGinit = 1
nTheta = 3
nEta = 3
nEps = 2
THETAinit = 2, 50, 0.1
OMinit = 0.2, 0.1, 0.1, 0.1, 0.2, 0.1, 0.1, 0.1, 0.2
SGinit = 0.1, 0, 0, 0.1
THETAinit = 4, 50, 0.2
The other values are the same with those of Theoph (ZERO, CONC).
[1] K. Bae. nmw: Understanding Nonlinear Mixed Effects Modeling for Population Pharmacokinetics. R package version 0.1.4. 2018. <URL: https://CRAN.R-project.org/package=nmw>.
[2] K. Bae and D. Yim. "R-based reproduction of the estimation process hidden behind NONMEM® Part 2: First-order conditional estimation". In: Translational and Clinical Pharmacology 24.4 (2016), p. 161. DOI: 10.12793/tcp.2016.24.4.161. <URL: https://doi.org/10.12793/tcp.2016.24.4.161>.
[3] M. Kim, D. Yim and K. Bae. "R-based reproduction of the estimation process hidden behind NONMEM® Part 1: first-order approximation method". In: Translational and Clinical Pharmacology 23.1 (2015), p. 1. DOI: 10.12793/tcp.2015.23.1.1. <URL: https://doi.org/10.12793/tcp.2015.23.1.1>.