Plotting optimized output seems to used wrong times

[krimobe]

When trying to visualize the optimized sample times on top of the predicted curve, this points seem to be off. To me it looks like a problem with the times.

`library(PopED)
library(deSolve)

sfg <- function(x,a,bpop,b,bocc){
parameters=c(CL=bpop[1]*exp(b[1]),
V=bpop[2]*exp(b[2]),
KA=bpop[3]*exp(b[3]),
Favail=bpop[4],
DOSE=a[1])
return(parameters)
}

system("R CMD SHLIB one_comp_oral_CL.c")
dyn.load(paste("one_comp_oral_CL", .Platform$dynlib.ext, sep = ""))

ff.ODE.compiled <- function(model_switch,xt,parameters,poped.db){
with(as.list(parameters),{
A_ini <- c(A1 = DOSE*Favail, A2 = 0)
times <- drop(xt)##xt[,,drop=T]
times <- sort(times)
times <- c(0,times) ## add extra time for start of integration
out <- ode(A_ini, times, func = "derivs", parms = c(CL,V,KA),
#jacfunc = "jac", # not really needed, speed up is minimal if this is defined or not.
dllname = "one_comp_oral_CL",
initfunc = "initmod", nout = 1, outnames = "Sum")
y = out[,"A2"]/(V)
y=y[-1] # remove initial time for start of integration
y = cbind(y) # must be a column matrix
return(list( y= y,poped.db=poped.db))
})
}

-- Define initial design and design space

poped.db.compiled <- create.poped.database(ff_fun=ff.ODE.compiled,
fg_fun = sfg,
fError_fun = feps.add.prop,
bpop=c(CL=0.15, V=8, KA=1.0, Favail=1),
notfixed_bpop=c(1,1,1,0),
d=c(CL=0.07, V=0.02, KA=0.6),
sigma=c(0.01,0.25),
groupsize=32,
xt=c( 0.5,1,2,6,24,36,72,120),
minxt=0,
maxxt=120,
a=c(DOSE=70),
mina=c(DOSE=0),
maxa=c(DOSE=100))

create plot of model with variability

plot_model_prediction(poped.db.compiled,IPRED=T,DV=T)

making optimization times more resonable

output <- poped_optim(poped.db.compiled ,opt_xt=T, opt_a=T, parallel=T, dlls = c('one_comp_oral_CL'))

plot_model_prediction(output$poped.db,IPRED=T,DV=T)`

Plot before optimization

Plot after optimization

[andrewhooker]

Hi.

It seems the ff.ode.compiled function is a bit outdated and is doing a strange filtering. I would change it to

ff.ODE.compiled <- function(model_switch,xt,parameters,poped.db){
  with(as.list(parameters),{
    A_ini <- c(A1 = DOSE*Favail, A2 = 0)
    times_xt <- drop(xt) #xt[,,drop=T] 
    times <- sort(times_xt)
    times <- c(0,times) ## add extra time for start of integration
    out <- ode(A_ini, times, func = "derivs", parms = c(CL,V,KA),
               #jacfunc = "jac", # not really needed, speed up is minimal if this is defined or not.
               dllname = "one_comp_oral_CL",
               initfunc = "initmod", nout = 1, outnames = "Sum")
    y = out[, "A2"]/(V)
    y=y[match(times_xt,out[,"time"])]
    y = cbind(y) # must be a column matrix
    return(list( y= y,poped.db=poped.db))
  })
} 

Which gives the plot after optimization of :