Least-Squares Regression Models interface

Use one of the following to add the object to a Workflow:

Right-click menu for a Workflow object: New > Least Squares Regression Models > [object name].
Main menu: Insert > Least Squares Regression Models > [object name].
Right-click menu for a worksheet: Send To > Least Squares Regression Models > [object name].

Note:To view the object in its own window, select it in the Object Browser and double-click it or press ENTER. All instructions for setting up and execution are the same whether the object is viewed in its own window or in Phoenix view.

Main Mappings panel
Dosing panel
Initial Estimates panel
PK Parameters panel
Units panel
Stripping Dose panel
Constants panel
Format panel
Linked Model tab
Weighting/Dosing Options tab
Parameter Options tab
Engine Settings tab
Plots tab (See the “Plots tab” description in the NCA section.)

Main Mappings panel

Use the Main Mappings panel to identify how input variables are used in a model. A separate analysis is performed for each profile. Required input is highlighted orange in the interface.

None: Data types mapped to this context are not included in any analysis or output.

Sort: Categorical variable(s) identifying individual data profiles, such as subject ID and treatment in a crossover study. A separate analysis is done for each unique combination of sort variable values.

X: For Dissolution, Indirect Response, Linear, PD, and PK/PD Linked Models. The independent vari­able in a dataset.

Y: For Dissolution, Indirect Response, Linear, PD, and PK/PD Linked Models. The dependent variable a dataset. Not needed if the Simulation checkbox is selected.

Time: For Michaelis-Menten, PK, and ASCII Models. Nominal or actual time collection points in a study.

Concentration: For Michaelis-Menten, PK Models, and ASCII Models. Drug concentration values in the blood. Not needed if the Simulation checkbox is selected.

Function: For ASCII Models. Model function.

Carry: Data variable(s) to include in the output worksheets. Note that time-dependent data variables (those that change over the course of a profile) are not carried over to time-independent output (e.g., Final Parameters), only to time-dependent output (e.g., Summary).

Weight: For Indirect Response, PK, PD, and PK/PD Linked Models. Used if weighting data is con­tained in the dataset. This column is only available if User Defined and Source are selected in the Weighting/Dosing Options tab.

Note:When using a PK operational object, external worksheets for stripping dose, units and initial esti­mates can be accessed in different ways. The differences will occur if there is more than 1 row of information on these external worksheets that correspond to one or more individual profiles of data of the Main input worksheet. In such cases, the stripping dose for PK models will be determined as the first value found on that external worksheet whereas the units and initial parameters will be based on the last row found on those external worksheets (for any given profile). To avoid any confusion stemming from these differences, it is suggested that external worksheets maintain a one-to-one row-based correspondence to the Main input profiles whenever possible.

Dosing panel

Available for Indirect Response, PK, and PK/PD Linked Models only, the Dosing panel allows users to type or map dosing data for the different models. The Dosing panel mapping columns change depending on the PK model type selected in the Model Selection tab. Required input is highlighted orange in the interface.

If multiple sort variables have been mapped in the Main Mappings panel, the Select sorts dialog is displayed so that the user can select the sort variables to include in the internal worksheet. Required input is highlighted orange in the interface.

None: Data types mapped to this context are not included in any analysis or output.

Sort: Sort variables.

Time: The time of dose administration.

Dose: The amount of drug administered. Dosing units are used.

End Time: The end time of the infusion.

Infusion Length: The total amount of time for an IV infusion. Only used in conjunction with a bolus dose.

Bolus: Amount of the bolus dose.

Amount Infused: Total amount of drug infused. Dosing units are used.

Note:If the dosing data is entered using the internal dosing worksheet, and different profiles require dif­ferent numbers of doses, then leave the Time, End Time, Dose, Infusion Length, Bolus, or Amount Infused cells blank for profiles that require less than the highest number of doses.

When using an internal worksheet, click the Rebuild button to reset the worksheet to its default state and delete all entered values.

Initial Estimates panel

The Initial Estimates panel allows users to type or map initial values and lower and upper boundaries for different indirect response parameters.

Note:Multiple-dose datasets require users to provide initial parameter values. For more on setting initial parameter estimates, see the “Parameter Estimates and Boundaries Rules” section.

If initial estimates and parameter boundaries for ASCII models are not set in the code, then they must be set in the Initial Estimates panel.

In datasets containing multiple sort variables, the initial estimates must be provided for each level of the sort variables unless the Propagate Final Estimates checkbox is selected. Checking this box applies the same types of parameter calculations and boundaries to all sort levels. Required input is highlighted orange in the interface.

If multiple sort variables have been mapped in the Main Mappings panel, the Select sorts dialog is displayed so that the user can select the sort variables to include in the internal worksheet.

None: Data types mapped to this context are not included in any analysis or output.

Sort: Sort variables.

Parameter: Model parameters such as v (volume) or km (Michaelis constant).

Initial: Initial parameter estimate values.

Fixed or Estimated: For Dissolution models. Whether the initial parameter is fixed or estimated.

Lower: Lower parameter boundary.

Upper: Upper parameter boundary.

The lower and upper values limit the range of values the parameters can take on during model fit­ting. This can be very valuable if the parameter values become unrealistic or the model will not converge. Although bounds are not always required, it is recommended that Lower and Upper bounds be used routinely. If they are used, every parameter must be given lower and upper bounds.

The Phoenix default bounds are zero for one bound and ten times the initial estimate for the other bound. For models with parameters that may be either positive or negative, user-defined bounds are preferred.

Note:For the WinNonlin Generated Initial Parameter Values option with Dissolution models, to avoid getting pop-up warnings that “WinNonlin will determine initial estimate” when using the Initial Esti­mates internal worksheet setup, delete the initial values and change the menu option from Esti­mated to Fixed before entering the initial estimates. Once the dropdown is changed from Fixed to Estimated, the initial value entered cannot be deleted and the warning pop-up will be displayed. Note that this situation does not affect the estimation process, as the entered initial value will not be used and WinNonlin will estimate the initial value as requested.

When using an internal worksheet, click the Rebuild button to reset the worksheet to its default state and delete all entered values.

PK Parameters panel

This panel is available for Indirect Response and PK/PD Linked models only.

Note:Users are required to enter initial PK parameter values in the PK Parameters panel in order for the model to run.

To display the units in the PK Parameters panel, units must be included in the time, concentration, and dose input data and the concentration units entered in the PK Units text field in the Model Selection tab.

If multiple sort variables have been mapped in the Main Mappings panel, the Select sorts dialog is dis­played so that the user can select the sort variables to include in the internal worksheet. Required input is highlighted orange in the interface.

None: Data types mapped to this context are not included in any analysis or output.
Sort: Sort variables.
Parameter: Name of the parameter.
Value: Value of the parameter.

When using an internal worksheet, click the Rebuild button to reset the worksheet to its default state and delete all entered values.

Units panel

For all Least-Squares Regression Models except Michaelis-Menten, an object’s display units can be changed to fit a user’s preferences. Required input is highlighted orange in the interface.

Depending on the type of model, there are some prerequisites for setting preferred units:

For Indirect Response and PK Models, the Time, Concentration, and Dose data must all contain units before users can set preferred units.

For PD and PK/PD Linked Models, the data mapped to the X and Y contexts must all contain units before users can set preferred units.

For ASCII Models, units must be set in the ASCII model code.

Each parameter used in a model and the parameter’s default units are listed in the Units panel.

None: Data types mapped to this context are not included in any analysis or output.
Name: Model parameters associated with the units.
Default: The model object’s default units.
Preferred: The user’s preferred units for the parameter.

When using an internal worksheet, click the Rebuild button to reset the worksheet to its default state and delete all entered values.

Note:if you see an “Insufficient units” message in the table, check that units are defined for time and concentration in your input.

Stripping Dose panel

The Stripping Dose panel is available when one of the macro constant PK models that use a stripping dose is specified in the Model Selection tab (i.e., models 8, 13, 14, 17, or 18). This panel is used to enter the stripping dose amount, which is the dose associated with initial parameter values for macro constant models.

If a user selects a macro constant PK model and provides user-specified initial estimates, then the user must specify the associated dose. If a user chooses to have Phoenix generate initial parameter values, then the stripping dose is identical to the administered dose.

None: Data types mapped to this context are not included in any analysis or output.
Sort: Sort variables.
Dose: The stripping dose.

If multiple sort values have been mapped in the Main Mapping panel, the Select sorts dialog is dis­played so that the user can select the sort variables to include in the internal worksheet.

Constants panel

The Constants panel is available only for Michaelis-Menten and ASCII Models and allows users to type or map dosing data for the model. Specific dosing information is determined using dosing con­stants. For more on how constants relate to dosing in Michaelis-Menten models, see “Dosing con­stants for the Michaelis-Menten model”. For more on how constants relate to dosing in ASCII models, see “ASCII Model dosing constants”.

When using an external worksheet for Constants, the Number of Constants will be determined from the number of rows per profile after mapping the following:

None: Data types mapped to this context are not included in any analysis or output.
Sort: Sort variables.
Order: The number of dosing constants used.
Value: The value for each dosing constant. For example, the value for CON[0] is 1 if the model is sin­gle-dose.

For an internal worksheet, set the Number of Constants on the Options panel, or specify NCON in the model text, to expand the internal worksheet for entering the Constants.

Format panel

The Format panel is only available for ASCII Models and is used to map ASCII code to an ASCII model. Users can view and edit ASCII model code in this panel.

Click Yes to have the ASCII code copied to the internal text editor.

Click No to remove the ASCII code and start with a blank internal text editor.

If ASCII code was previously mapped to the User ASCII Model panel then that code is displayed in the internal text editor, instead of a blank panel.

More information on the text editor is available on the Syncfusion website.


Last modified date:7/9/20
Certara USA, Inc.
Legal Notice | Contact Certara
© 2020 Certara USA, Inc. All rights reserved.