The Maximum Likelihood Models object (also referred to as NLME) can perform a variety of pharmacokinetic and pharmacodynamic analyses using individual and population modeling. It provides access to robust and efficient Maximum Likelihood engines to perform individual, population, and pooled data analyses. NLME provides automated covariate selection, bootstrap, and visual predictive check options for population models. It creates consistent graphical and worksheet output to allow easy comparison between models via the Maximum Likelihood Model Comparer object.
Note: Phoenix NLME is only supported on 64-bit systems.
NLME is extremely flexible on requirements for input data. There are no requirements for naming a variable as long as it is acceptable in a Phoenix spreadsheet (which does not allow special characters) and is not a reserved word for Phoenix NLME (see the “Reserved and user-defined variable names” section in the PML documentation). Column headers can contain any combination of alphanumeric characters and underscores.
There are three ways of creating a custom PK model.
The built-in model interface uses the menus in the Maximum Likelihood Models object to create a model, which is customizable to the extent that various options and selections may be combined as desired. See the “Object and built-in model interface” section for more information.
The Graphical model interface uses the graphical model editor to create the model structure in diagram form.
The Textual model interface is for entering a custom model using PML (Pharmacometrics Modeling Language), which allows for the greatest amount of flexibility in model structure and customization.
Combinations of these three methods may be necessary to build the desired model. For example, when preparing a recycling model (enterohepatic recirculation), a graphical model can be built but textual changes must be made to complete the model, such as:
Add a function, called switch, to turn on and off the recycling process.
double(Switch)
Adjust the structural model equations to use the Rate variable.
deriv(Abile=(A1*K1g)-(Abile*Rate))
deriv(Agut=(Abile*Rate)-(Agut*Ka))
Rate=Switch/Tau #Tau is the gall bladder emptying interval
Add a sequence block to define the gall bladder emptying process.
Ri=10 #Time recirculation occurs, 10hrs used here
sequence{
Switch=0; #Turn off the gall bladder emptying
sleep(Ri); #Wait for 10 hours
Switch=1; #Turn on the gall bladder emptying on
sleep(Tau); #Wait for the gall bladder emptying interval
Switch=0; #Turn off the gall bladder emptying
}
Add fixed effect for Tau.
fixef(Tau=c(, 3,))
Comment out any unused parameters and adjust fixed effect values as needed.
Note: In the NLME interface, where numbers can be entered in data fields, generally either comma (,) or period (.) can be used as a decimal point. (It will be converted to a period.) However, there are fields where sequences of numbers, separated by commas, can be entered, such as the sequence of times in a table specification. In those fields, the comma character cannot be used as a decimal point, because it acts as a delimiter between numbers.
The following symbol in the Phoenix NLME documentation indicates information specific to individual modeling,
This section contains information on the following topics:
Object and built-in model interface
Job control for parallel and remote execution
Differential Equations in NLME