Testing the Phoenix installation

The following instructions test the installation of Phoenix WinNonlin by using a number of the sample files provided with the software. This chapter is not intended as a full validation of the product. It is intended to test for proper installation of major components of the application.

A set of Validation Suite Products is available from Certara. Contact the Certara sales department for more information.

Start Phoenix and create a new project

  1. Double-click the Phoenix icon (Phoenix_icon_1.png) on your desktop to start Phoenix.

  2. Note:AutoPilot users running on 64-bit operating systems should run the 32-bit version of Phoenix (Phoenix32) as opposed to the 64-bit version. The Phoenix desktop icon should be updated to ref­erence the Phoenix32 executable as well.

  3. Select File > New Project to create a new project. A new project is created in the Object Browser.

  4. Name the new project Install Test.

The left panel’s default view is the Object Browser, which contains the project, and the other folders and objects that are contained in the project. The right viewing panel’s default view is blank, unless one of the project folders or the workflow is selected.

phxdefaultnewscreen.png 

Creation of Install Test project denotes successful installation test

Confirm license installation

The ability to execute different Phoenix plug-ins depends on the license type installed. Use the Prefer­ences dialog to check the installed Phoenix license(s).

  1. Select Edit > Preferences to access the Preferences dialog.

  2. Select Licensing > License Management to access the list of available licenses.

  3. Confirm that the purchased license or licenses are listed in the Licenses panel.

Confirm plug-in startup

The Phoenix architecture is based on a series of plug-ins that allow different features to be enabled or disabled. Phoenix initializes these plug-ins when the application starts. The default setting is for all plug-ins to initialize when Phoenix starts. The plug-in initialization is independent of the installed license(s).

  1. In the Preferences dialog, click Plugins to view the list of available plug-ins.

  2. The Plugins panel on the right contains three tabs: General, System, and Non-loaded.

    Each plug-in has two possible states, Started and Stopped.

    All plug-ins are started the first time Phoenix is started.

    phxpreferencesplugins.png 

    Plugins Preferences panel

  3. Select the General, System, and Non-loaded tabs to see the review the state of each plug-in type and to confirm they are started.

  4. Click OK to exit the Preferences dialog.

Import a dataset

The dataset Bguide1.dat is used to test key Phoenix functions.

  1. Select File > Import, click icon_importfile_2.png (Import File icon), or right-click the Data Folder in the Object Browser and select Import to display the Import File(s) dialog.

  2. Navigate to <Phoenix_install_dir>\application\Examples\WinNonlin\Sup­porting files).

  3. Select the file Bguide1.dat and click Open.
    The File Import Wizard dialog is displayed. The dialog is used to assign options for how the data are imported and presented.

  4. Click Finish. The dataset is added to the project Data folder and the worksheet is displayed in the right viewing panel.

phoenix_view_worksheet.png 

View of a dataset

Create a plot

  1. Select the Workflow object in the Object Browser and then select Insert > Plotting > XY Plot.
    Or
    Right-click the Workflow object and select New > Plotting > XY Plot.

    The XY Plot object is added to the workflow in the Object Browser and is automatically opened in the right viewing panel. The default view of an object is the Setup tab, which contains all the steps necessary to set up an object.

  2. Use the pointer to drag the Bguide1 worksheet from the Data folder to the XY Data Mappings panel.

  3. Use the option buttons in the XY Data Mappings panel to map the data types to the following con­texts:
    Map Subject to the Group context.
    Map Time to the X context.
    Map Conc to the Y context.

  4. phxxymapping.png 

    Context mapping for XY Plot

  5. Click icon_execute_42.png (Execute icon) to execute the workflow.

  6. XYPlotresults.png 

    XY Plot results

  7. Now use the Bguide1 dataset to test the Table object and its summary statistics function.

Create a table

  1. Add the Table object to the workflow by selecting the Workflow object in the Object Browser and then selecting Insert > Reporting > Table.

  2. Map the dataset Bguide1 as the input source for the Table object by dragging the Bguide1 work­sheet from the Data folder to the Table object Main Mappings panel.

  3. Use the option buttons in the Main Mappings panel to map the data types to the following con­texts:
    Map Subject to the Stratification Row context.
    Leave Time mapped to None.
    Map Conc to the Data context.

  4. In the Options tab (located below the Setup tab, specify which table type the Table object is to use by selecting Table 1 - Column Summary by Row Stratification from the Table Type menu.

  5. Select the Page Break on Row Stratification checkbox.

  6. phxtablemapping.png 

    Setting Table object options

  7. Select the Statistics tab, which is located underneath the Setup tab.

  8. Click Select All to select all output statistics.

  9. Execute the object.

The results are presented as three HTML tables in the Results tab. Compare the tables in the Results tab to the tables pictured below.

testinstall00654.png 

Resulting HTML table for subject DW

quicktour_tables_gs.png 

Resulting HTML table for subject GS

quicktour_tables_rh.png 

Resulting HTML table for subject RH

Confirm noncompartmental analysis functions

  1. Select File > Load Project. The Load Project dialog is displayed.

  2. Navigate to <Phoenix_install_dir>\application\Examples\WinNonlin.

  3. Select Multiple_Profiles.phxproj and click Open.

    This project contains:

  4. A dataset worksheet (profiles)

    A worksheet of dosing information (Dosing published from NCA)

    An XY Plot object

    An NCA model object

    A Descriptive Statistics object

    A Data Wizard object

    An X-Categorical XY Plot object

  5. Expand the workflow node.

  6. Select the NCA model object in the Object Browser.

  7. Select items in the Setup tab list to explore the data mappings and option settings.

  8. Execute the object.

Text output

The Core output contains the model settings and the same data as the worksheets, but presented in plain ASCII text. If there were errors in the model they would be listed here. Part of this file is shown below:

  …
  Model: Plasma Data, Extravascular Administration
  Number of nonmissing observations:   12
  Dose time:      0.00
  Dose amount:      100.00
  Calculation method: Linear Trapezoidal with Linear Interpolation
  Weighting for lambda_z calculations: Uniform weighting
  Lambda_z method: Find best fit for lambda_z, Log regression
  Compute Concentrations at: 75
  
  Summary Table
  -------------
   Time  Conc.  Pred. Residual    AUC         AUMC      Weight
   min   ng/ml  ng/ml  ng/ml   min*ng/ml  min*min*ng/ml
  ------------------------------------------------------------
  0.0000 0.0000                   0.0000       0.0000
  5.000  1816.                     4540.    2.270e+04
  10.00  1687.                 1.330e+04    8.756e+04
  15.00* 286.8  327.4 -40.61   1.823e+04    1.405e+05   1.000
  20.00* 323.7  311.1  12.65   1.976e+04    1.674e+05   1.000
  30.00* 321.5  280.8  40.75   2.290e+04    2.480e+05   1.000
  45.00* 237.3  240.7 -3.445   2.717e+04    4.004e+05   1.000
  60.00* 218.0  206.4  11.56   3.059e+04    5.786e+06   1.000
  90.00* 192.6  151.8  40.80   3.675e+04    1.035e+06   1.000
  120.0* 123.8  111.6  12.18   4.149e+04    1.518e+06   1.000
  180.0* 40.00  60.35 -20.35   4.641e+04    2.179e+06   1.000
  240.0* 36.20  32.63  3.572   4.869e+04    2.656e+06   1.000
 360.0* 5.500  9.538 -4.038   5.119e+04    3.296e+06   1.000
 480.0* 4.300  2.788  1.512   5.178e+04    3.539e+06   1.000

The Settings file lists all the settings used to specify the noncompartmental analysis.

  …
    Sort: Subject, Form
    Time: Time [min]
    Concentration: Conc [ng/mL]
    Carry:
  Dosing: (Internal)
  Slopes: (Internal)
  Partial Areas: (Internal)
 Therapeutic Response: <None>
  Units: (Internal)
 Parameter Names: <None>
  …
  Plasma Model
  Title=Processing Multiple Profiles with Model 200
  Linear Trapezoidal Linear Interpolation
  Sparse=False
  Weighting=Uniform Weighting; 0
  Dose Type=Extravascular
  Dose Unit=ng
  Dose Normalization=None
  Compute Concentrations at: 75

Output data

The NCA object creates seven results worksheets: Dosing Used, Exclusions, Final Parameters, Final Parameters Pivoted, Partial Areas, Plot Titles, Slopes Settings, and Summary Table. Selections from the Final Parameters and Summary Table worksheets are shown below.

NCAFinalParsub1form1.png 

Final Parameters worksheet

NCASummaryTablesub1.png 

Summary Table worksheet

Plots

A total of 12 plots are generated; one for each of two formulations, for each of the six subjects. The first two charts for subject one are shown below.

NCAplotsub1form1_2.png 

Plots for subject one, capsule and tablet formulation

Confirm pharmacokinetic modeling functions

  1. Select File > Load Project. The Load Project dialog is displayed.

  2. Navigate to <Phoenix_install_dir>\application\Examples\WinNonlin.

  3. Select PK_Model.phxproj and click Open.

    This project contains:

  4. A dataset worksheet (study1).

    An XY Plot object.

    A PK model object.

  5. Expand the workflow node.

  6. Select the PK Model object in the Object Browser.

  7. Select items in the Setup tab list to explore the model’s data mappings and option settings.

    The imported PK Model object uses PK Model 3, which is a one-compartment model with 1st order absorption.

  8. Execute the object.

Worksheet results

The PK Model object’s output worksheets partially include Condition Numbers, Diagnostics, Dosing Used, Final Parameters, Initial Estimates, Secondary Parameters, and Summary Table. The Final Parameters, Secondary Parameters, and Summary Table worksheets are shown below.

PKmodelFinalParams.png 

Final Parameters worksheet

PKmodelSecondaryParams.png 

Secondary Parameters worksheet

PKmodelSummaryTable.png 

Summary Table worksheet

Text output

The Core output text results include all model settings and iterations, including the output from the worksheets. Any model errors would be listed here. Part of this file is shown below.

  …
 Listing of input commands
  MODEL 3
  NVAR 3
  NPOI 1000
  XNUM 2
  YNUM 3
  NCON 3
  CONS 1,2,0
  METH 2‘Gauss-Newton (Levenberg and Hartley)
  ITER 50
  INIT 0.25,1.81,0.23
  MISS ‘.’
  DATA ‘WINNLIN.DAT’
  BEGIN

The Settings file lists all the settings used to specify the noncompartmental analysis. Part of this file is shown below.

  …
  Main: PK Model.Data.study1
    Sort: Subject
    Time: Time [hr]
    Concentration: Conc [ng/mL]
    Carry:
  Dosing: (Internal)
  Initial Estimates: (Internal)
  Units: (Internal)
  ***** Other Parameters *****
  …
  PK 3-[PK]
  Gauss-Newton (Levenberg and Hartley)
  Convergence criteria of 0.0001 used during minimization process
  50 maximum iterations allowed during minimization process

Plots

The plot results include Observed Y and Predicted Y vs X, Partial Derivatives Plot, Predicted Y vs Observed Y, Predicted Y vs X, Residual Y vs Predicted Y, and Residual Y vs X. Some plot results are shown below.

pkobsypredyvsx.png 

Observed Y and Predicted Y vs X

pkmodelPredYvsObsY_2.png 

Predicted Y vs Observed Y

pkmodelResidualYvsX_2.png 

Residual Y vs X

Confirm bioequivalence functions

  1. Select the Install Test project in the Object Browser.

  2. Select File > Import.

  3. In the Import File(s) dialog, navigate to <Phoenix_install_dir>\application\Exam­ples\WinNonlin\Supporting files.

  4. Select the file Seq2Per4.csv and click Open.

  5. In the File Import Wizard dialog, click Finish. The dataset is added to the project Data folder.

  6. Select the project workflow in the Object Browser and then select Insert > Computation Tools > Bioequivalence.
    The Bioequivalence object is added to the workflow in the Object Browser.

  7. Note:The default settings for a new Bioequivalence model are Crossover as the type of study and Average as the type of bioequivalence.

  8. Use the pointer to drag the Seq2Per4 worksheet from the Data folder to the Main Mappings panel.
    The Seq2Per4 dataset is mapped to the Bioequivalence object.

  9. Use the option button in the Main Mappings panel to map AUC to the Dependent context.

    The following data types are automatically mapped to contexts when the dataset is mapped to the Bioequivalence model.
    Sequence is mapped to the Sequence context.
    Subject is mapped to the Subject context.
    Period is mapped to the Period context.
    Formulation is mapped to the Formulation context.
    AUC is mapped to the Dependent context.

  10. In the Model tab (located below the Setup tab), ensure that Crossover is selected as the Type of study, Average is selected as the Type of Bioequivalence, and R is selected as the Reference Formulation.

  11. phxbiomapping.png 

    Settings for a Bioequivalence model

  12. Select the Fixed Effects tab, which is located below the Setup tab.
    Ln(x) is automatically selected in the Dependent Variables Transformation menu. Do not change this setting.

  13. Execute the object.

Output data

The bioequivalence model worksheet output partially includes Average Bioequivalence, Diagnostics, Final Fixed Parameters, Final and Initial Variance Parameters, Least Squares Means, and Sequential Tests. The Diagnostics, Final Variance Parameters, and Sequential Tests worksheets are shown below.

phxbioavebio_1.png 

Average Bioequivalence worksheet

BioEq1Diagnostics_1.png 

Diagnostics worksheet

BioEq1FinalVarParams_1.png 

Final Variance Parameters worksheet

BioEq1SequentialTests_1.png 

Sequential Tests worksheet

This concludes the Phoenix installation tests.

Test installation of Phoenix NLME

This section tests Phoenix WinNonlin’s population modeling functions. This test is not intended as a full validation of population modeling.

Users who have purchased a license for Phoenix NLME can use the Maximum Likelihood Models object to create and test population models. After Phoenix is installed, follow the steps below to con­firm that population modeling is working correctly.

If any problems are encountered during this installation test, report the issue to Customer Support.

Start Phoenix and create a project

  1. Double-click the Phoenix icon (Phoenix_icon_2.png) on your desktop to start Phoenix.

  2. Select File > New Project to create a new project. A new project is created in the Object Browser.

  3. Name the new project Installation Project.

  4. To verify active licenses select Help > About Phoenix. For this installation test to work users must have an active license for Phoenix NLME 8.1.

Confirm population modeling functionality

This example verifies that the Phoenix NLME license and the Minimal GNU C++ compiler are cor­rectly installed and configured. If Phoenix is being used without population modeling functionality, then skip this part of the installation test.

  1. Select File > Import.

  2. In the Import File(s) dialog, navigate to <Phoenix_install_dir>\application\Exam­ples\NLME\Supporting files.

  3. Select theopp.csv and click Open.

  4. In the File Import Wizard dialog, click Finish. The dataset is added to the project Data folder and the worksheet is opened.

  5. Select the workflow object in the Object Browser.

  6. Select File > Load Workflow Template.

  7. In the dialog, navigate to <Phoenix_install_dir>\application\Exam­ples\NLME\Supporting files.

  8. Select Diagonal Model.phxtmplt and click Open.

    WinNonlin template files store information about settings and internal worksheets, but do not store external data sources.

    The template file adds a one-compartment, 1st order absorption model that uses weight as a covariate. The model uses the FOCE L-B (First Order Conditional Estimation with the Lindstrom-Bates iterative algorithm) run method.

  9. Use the pointer to drag the theopp worksheet from the Data folder to the Main Mappings panel.

    The theopp dataset is mapped to the Diagonal model and the study variables xid, dose, time, yobs, and wt are automatically mapped to the contexts ID, Aa, Time, CObs, and wt, respectively.

  10. Execute the object.

  11. Close the NLME Job Status dialog by clicking the X in the upper right corner.

Verify the results

  1. Select the Verification tab.

    The Verification Results state that the model successfully executed, and that no errors occurred.

  2. poppkVerificationtab.png 

    Verification tab

  3. Select the Results tab.

The model object produces text files, plots, and worksheets. The output types are listed below.

Output data

The model creates 23 worksheet results. Some of the output worksheets are Eta, EtaCovariate, Omega, and Theta. Not all worksheets listed under Output Data contain results.

Confirm the accuracy of some of the output in order to complete the installation test.

  1. In the Output Data, select the Theta worksheet and confirm that the following estimates are the same (after rounding):
    tvKa     1.55696
    tvV       0.455554
    tvCl      0.0402882
    stdev0  0.692202

poppkThetaworksheet.png 

Theta worksheet

poppkEtaworksheet.png 

Eta worksheet

Plots

Thirty-nine plots are listed in the Plots section. Not all listed plots have results.

  1. In the Plot output, select Pop CWRES vs PRED and compare it to the screen shot below:

poppkPopCWRESvsPRED.png 

Pop CWRES vs PRED

Text output

The Maximum Likelihood Models object creates six text files. The Settings text file lists model set­tings, part of which is shown below.

  …
    Sort:
    ID: xid
    Aa: dose
    Time: time
    CObs: yobs
    wt: wt
  …
 test(){
    deriv(Aa = -Ka * Aa)
    deriv(A1 = Ka * Aa - Cl * C)
    dosepoint(Aa)
    C=A1/V
    error(CEps = 1)
    observe(CObs = C + CEps)
    stparm(Ka = tvKa * exp(nKa))
    stparm(V = tvV * exp(nV))
    stparm(Cl = tvCl * exp(nCl))
    covariate(wt)
    fixef(tvKa = c(, 1.54697,))
    fixef(tvV = c(, 0.455465,))
    fixef(tvCl = c(, 0.0402979,))
    ranef(diag(nV, nKa, nCl)=c(0.01809, 0.41265, 0.06997))
  }
  ------------------------------------
  id("xid")
  time("time")
  dose(Aa<-"dose")
  covr(wt<-"wt")
  obs(CObs<-"yobs")
  table(file=”posthoc.csv”, time(0), Ka,V,Cl)
  ------------------------------------
  Run Options
  Method: FOCE L-B
  N Iter:1000
  Input sorted by subject+time
 Enabling automatic log transform (if applicable)
  ODE solver method: matrix exponent
  Method of computing standard errors: Central Diff
  Hessian standard errors
  Confidence Level %95
  Simple run was performed

This concludes the installation test for Phoenix NLME. Close project without saving it.


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