mattfidler
commented
5 years ago Hi @motyocska,
You have to have ctime in the solving dataset or specify it directly in the covariate In the example you referred to, ctime is in a separate dataset as a time-varying covariate.
library(RxODE)
mod3 <- RxODE({
KA=2.94E-01;
CL=1.86E+01;
V2=4.02E+01;
Q=1.05E+01;
V3=2.97E+02;
Kin0=1;
Kout=1;
EC50=200;
## The linCmt() picks up the variables from above
C2 = linCmt();
Tz= 8
amp=0.1
eff(0) = 1 ## This specifies that the effect compartment starts at 1.
## Kin changes based on time of day (like cortosol)
Kin = Kin0 +amp *cos(2*pi*(ctime-Tz)/24)
d/dt(eff) = Kin - Kout*(1-C2/(EC50+C2))*eff;
})
ev <- eventTable(amount.units="mg", time.units="hours") %>%
add.dosing(dose=10000, nbr.doses=1, dosing.to=2) %>%
add.sampling(seq(0,48,length.out=100));
## Create data frame of 8 am dosing for the first dose
cov.df <- data.frame(ctime =(seq(0,48,length.out=100)+8) %% 24);
Now there is a covariate present, the system can be solved using the cov option
## Solve with Next Observation Carried Backward (NONMEM's interpolation method)
r1 <- solve(mod3, ev, covs=cov.df,covsInterpolation="nocb")
which gives
r1
▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ Solved RxODE object ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂
── Parameters (r1$params): ─────────────────────────────────────────────────────
KA CL V2 Q V3 Kin0 Kout
0.294000 18.600000 40.200000 10.500000 297.000000 1.000000 1.000000
EC50 Tz amp pi
200.000000 8.000000 0.100000 3.141593
── Covariates (r1$covs): ───────────────────────────────────────────────────────
# A tibble: 100 x 1
ctime
<dbl>
1 8
2 8.48
3 8.97
4 9.45
5 9.94
6 10.4
# ... with 94 more rows
── Initial Conditions (r1$inits): ──────────────────────────────────────────────
eff
1
── First part of data (object): ────────────────────────────────────────────────
# A tibble: 100 x 4
time C2 Kin eff
<dbl> <dbl> <dbl> <dbl>
1 0 0 1.10 1
2 0.485 27.8 1.10 1.07
3 0.970 43.7 1.09 1.14
4 1.45 51.8 1.09 1.21
5 1.94 54.8 1.08 1.26
6 2.42 54.6 1.07 1.29
# ... with 94 more rows
▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂Another option is to have a dataset with both pieces in it
dat <- ev$get.EventTable() ## Get the event table dataset
dat$ctime <- c(8,seq(0,48,length.out=100)+8 %% 24)
## Solve with Last Observation Carried Forward (Monolix/nlmixr's interpolation method)
r1 <- solve(mod3, dat, covsInterpolation="locf")Which gives the same solution:
> r1
▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂ Solved RxODE object ▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂
── Parameters (r1$params): ─────────────────────────────────────────────────────
KA CL V2 Q V3 Kin0 Kout
0.294000 18.600000 40.200000 10.500000 297.000000 1.000000 1.000000
EC50 Tz amp pi
200.000000 8.000000 0.100000 3.141593
── Initial Conditions (r1$inits): ──────────────────────────────────────────────
eff
1
── First part of data (object): ────────────────────────────────────────────────
# A tibble: 100 x 4
time C2 Kin eff
<dbl> <dbl> <dbl> <dbl>
1 0 0 1.1 1
2 0.485 27.8 1.10 1.07
3 0.970 43.7 1.10 1.15
4 1.45 51.8 1.09 1.22
5 1.94 54.8 1.09 1.27
6 2.42 54.6 1.08 1.31
# ... with 94 more rows
▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂▂
motyocska
Matt,
how would one signify the current time point in the solver if the value of the time point itself is used in an equation within the model in and ODE model using nlmixr? I am looking for the FORTRAN equivalent of T (or t), or with an earlier example of yours from RxODE you may recall the time varying co variate issue with a model where you had an equation as this:
how would you refer to ctime as the current time point in the solver? I tried simply enter "time" but will not take it as seems to ask for it to be declared?
appreciate your help
Andras