In nlmixrdevelopment/RxODE.doc: Facilities for Simulating from ODE-Based Models

options(cli.unicode=FALSE, crayon.enabled=FALSE);
options(knitr.table.format = "html")
htmltools::img(src = knitr::image_uri("logo.png"), 
               alt = 'RxODE', 
               style = 'position:absolute; top:0; right:0; padding:10px; border: 0;')
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  message = FALSE,
  warning = FALSE,
  out.width = "100%"
)

# Using RxODE data frames
## Creating an interactive data frame

RxODE supports returning a solved object that is a modified data-frame.
This is done by the `predict()`, `solve()`, or `rxSolve()` methods.

```r

library(RxODE)
library(units)

## Setup example model
mod1 <-RxODE({
    C2 = centr/V2;
    C3 = peri/V3;
    d/dt(depot) =-KA*depot;
    d/dt(centr) = KA*depot - CL*C2 - Q*C2 + Q*C3;
    d/dt(peri)  =                    Q*C2 - Q*C3;
    d/dt(eff)  = Kin - Kout*(1-C2/(EC50+C2))*eff;
});

## Seup parameters and initial conditions

theta <- 
    c(KA=2.94E-01, CL=1.86E+01, V2=4.02E+01, # central 
      Q=1.05E+01,  V3=2.97E+02,              # peripheral
      Kin=1, Kout=1, EC50=200)               # effects

inits <- c(eff=1);

## Setup dosing event information
ev <- eventTable(amount.units="mg", time.units="hours") %>%
    add.dosing(dose=10000, nbr.doses=10, dosing.interval=12) %>%
    add.dosing(dose=20000, nbr.doses=5, start.time=120,dosing.interval=24) %>%
    add.sampling(0:240);


## Now solve
x <- predict(mod1,theta, ev, inits)
print(x)

or

x <- solve(mod1,theta, ev, inits)
print(x)

Or with mattigr

x <- mod1 %>% solve(theta, ev, inits)
print(x)

Using the solved object as a simple data frame

The solved object acts as a data.frame or tbl that can be filtered by dpylr. For example you could filter it easily.

library(dplyr)
## You can  drop units for comparisons and filtering
x <- mod1 %>% solve(theta,ev,inits) %>% drop_units %>% filter(time <= 3) %>% as.tbl
## or keep them and compare with the proper units.
x <- mod1 %>% solve(theta,ev,inits) %>% filter(time <= set_units(3, hr)) %>% as.tbl
x

Updating the data-set interactively

However it isn't just a simple data object. You can use the solved object to update parameters on the fly, or even change the sampling time.

First we need to recreate the original solved system:

x <- mod1 %>% solve(theta,ev,inits);
print(x)

Modifying initial conditions

To examine or change initial conditions, you can use the syntax cmt.0, cmt0, or cmt_0. In the case of the eff compartment defined by the model, this is:

x$eff0

which shows the initial condition of the effect compartment. If you wished to change this initial condition to 2, this can be done easily by:

x$eff0 <- 2
print(x)
plot(x)

Modifying observation times for RxODE

Notice that the initial effect is now 2.

You can also change the sampling times easily by this method by changing t or time. For example:

x$t <- seq(0,5,length.out=20)
print(x)
plot(x)

Modifying simulation parameters

You can also access or change parameters by the $ operator. For example, accessing KA can be done by:

x$KA

And you may change it by assigning it to a new value.

x$KA <- 1;
print(x)
plot(x)

You can access/change all the parameters, initialization(s) or events with the $params, $inits, $events accessor syntax, similar to what is used above.

This syntax makes it easy to update and explore the effect of various parameters on the solved object.

nlmixrdevelopment/RxODE.doc documentation built on Oct. 17, 2019, 1:27 a.m.