predict(LeastSquaresResults,OptimResults,NLINResults)

Simulate and evaluate fitted SimBiology model

Syntax

[ynew,parameterEstimates]= predict(resultsObj)
[ynew,parameterEstimates]= predict(resultsObj,data,dosing)

Description

[ynew,parameterEstimates]= predict(resultsObj) returns simulation results ynew and parameter estimates parameterEstimates of a fitted SimBiology® model.

[ynew,parameterEstimates]= predict(resultsObj,data,dosing) returns simulation results ynew and estimated parameter values parameterEstimates from evaluating the fitted SimBiology model using the specified data and dosing information.

During simulations, predict uses the parameter values in the resultsObj.ParameterEstimates property. Use this method when you want to evaluate the fitted model and predict responses using new data and/or dosing information.

Input Arguments

collapse all

Estimation results, specified as an OptimResults object or NLINResults object, which contains estimation results returned by sbiofit. It must be a scalar object.

Output times or grouped data, specified as a vector, or cell array of vectors of output times, or groupedData object.

If it is a vector of time points, predict simulates the model with new time points using the parameter estimates from the results object resultsObj.

If it is a cell array of vectors of time points, predict simulates the model n times using the output times from each time vector, where n is the length of data.

If it is a groupedData object, it must have an independent variable such as Time. It must also have a group variable if the training data used for fitting has such variable. You can use a groupedData object to query different combinations of categories if the resultsObj contains parameter estimates for each category. predict simulates the model for each group with the specified categories. For instance, suppose you have a set of parameter estimates for sex category (males versus females), and age category (young versus old) in your training data. You can use predict to simulate the responses of an old male (or any other combination) although such patient may not exist in the training data.

If the resultsObj is from estimating category-specific parameters, data must be a groupedData object.

Note

If UnitConversion is turned on for the underlying SimBiology model that was used for fitting and data is a groupedData object, data must specify valid variable units via data.Properties.VariableUnits property. If it is a numeric vector or cell array of vectors of time points, predict uses the model’s TimeUnits.

Dosing information, specified as an empty array [] or 2-D matrix of SimBiology dose objects (ScheduleDose object or RepeatDose object). If dosing is a matrix of dose objects, the matrix must contain default doses or be consistent with the original dosing data used with sbiofit. That is, dose objects in dosing must have the same values for dose properties (such as TargetName) or must be parameterized in the same way as the original dosing data. For instance, suppose that the original dosing matrix has two columns of doses, where the doses in the first column target species x and those in the second column target species y. Then dosing must have doses in the first column targeting species x and those in the second column targeting species y.

  • If empty, no doses are applied during simulation, even if the model has active doses.

  • If not empty, the matrix must have a single row or one row per group in the input data. If it has a single row, the same doses are applied to all groups during simulation. If it has multiple rows, each row is applied to a separate group, in the same order as the groups appear in the input data. If some groups (or time vectors) require more doses than others, then fill in the matrix with default (dummy) doses.

  • Multiple columns are allowed so that you can apply multiple dose objects to each group or time vector. All doses in a column must be default doses or must reference the same components in the model (for instance, the doses must have the same dose target TargetName) and must have consistent parameterized properties as in the original dosing data used with sbiofit. For example, if the Amount property of a dose used in the original sbiofit call is parameterized to a model-scoped parameter 'A', all doses for the corresponding group (column) in dosing must have the Amount property parameterized to 'A'.

  • A default dose has default values for all properties, except for the Name property. Create a default dose as follows.

    d1 = sbiodose('d1');

  • In addition to manually constructing dose objects using sbiodose, if the input data is a groupedData object and has dosing information, you can use the createDoses method to construct doses from it.

The number of rows in the dosing matrix and the number of groups or output time vectors in data determine the total number of simulation results in the output ynew. For details, see the table in the ynew argument description.

Note

If UnitConversion is turned on for the underlying SimBiology model that was used for fitting, dosing must specify valid amount and time units.

Output Arguments

collapse all

Simulation results, returned as a vector of SimData objects. The states reported in ynew are the states that were included in the responseMap input argument of sbiofit as well as any other states listed in the StatesToLog property of the runtime options (RuntimeOptions) of the SimBiology model.

The total number of simulation results in ynew depends on the number of groups or output time vectors in data and the number of rows in the dosing matrix.

Number of groups or output time vectors in dataNumber of rows in the dosing matrixSimulation results

1

0, that is, dosing is empty []

The total number of SimData objects in ynew is 1.

No doses are applied during simulation.

1

1

The total number of SimData objects in ynew is 1.

The given row of doses is applied during the simulation.

1

N

The total number of SimData objects in ynew is N.

Each row of dosing is applied to each simulation.

N

0, that is, dosing is empty []

The total number of SimData objects in ynew is N.

No doses are applied during simulation.

N

1

The total number of SimData objects in ynew is N.

The same row of doses is applied to each simulation.

NN

The total number of SimData objects in ynew is N.

Each row of dosing is applied to a separate group, in the same order that groups appear in data.

MNThe function throws an error when MN.

Estimated parameter values, returned as a table. This is identical to resultsObj.ParameterEstimates property. The predict method uses these parameter values during simulation.

Examples

collapse all

This example uses the yeast heterotrimeric G protein model and experimental data reported by [1]. For details about the model, see the Background section in Parameter Scanning, Parameter Estimation, and Sensitivity Analysis in the Yeast Heterotrimeric G Protein Cycle.

Load the G protein model.

sbioloadproject gprotein

Enter the experimental data containing the time course for the fraction of active G protein, as reported in the reference paper [1].

time = [0;10;30;60;110;210;300;450;600];
GaFracExpt = [0;0.35;0.4;0.36;0.39;0.33;0.24;0.17;0.2];

Create a groupedData object based on the experimental data.

tbl = table(time,GaFracExpt);
grpData = groupedData(tbl);

Map the appropriate model component to the experimental data. In other words, indicate which species in the model corresponds to which response variable in the data. In this example, map the model parameter GaFrac to the experimental data variable GaFracExpt from grpData.

responseMap = 'GaFrac = GaFracExpt';

Use an estimatedInfo object to define the model parameter kGd as a parameter to be estimated.

estimatedParam = estimatedInfo('kGd');

Perform the parameter estimation.

fitResult = sbiofit(m1,grpData,responseMap,estimatedParam);

View the estimated parameter value of kGd.

fitResult.ParameterEstimates
ans = 

    Name     Estimate    StandardError
    _____    ________    _____________

    'kGd'    0.11        3.683e-05    

Suppose you want to simulate the fitted model using different output times than those in the training data. You can use the predict method to do so.

Create a new variable T with different output times.

T = [0;10;50;80;100;150;300;350;400;450;500;550];

Use the predict method to simulate the fitted model on the new time points. No dosing was specified when you first ran sbiofit. Hence, you cannot use any dosing information with the predict method, and an empty array must be specified as the third input argument.

ynew = predict(fitResult,T,[]);

Plot the simulated data with the new output times.

sbioplot(ynew)

Expand Run1 and only select GaFrac to display its simulated data.

This example shows how to estimate category-specific (such as young versus old, male versus female) PK parameters using the profile data from multiple individuals using a two-compartment model. The parameters to estimate are the volumes of central and peripheral compartment, the clearance, and intercompartmental clearance.

The synthetic data used in this example contains the time course of plasma concentrations of multiple individuals after a bolus dose (100 mg) measured at different times for both central and peripheral compartments. It also contains categorical variables, namely Sex and Age.

clear
load(fullfile(matlabroot,'examples','simbio','sd5_302RAgeSex.mat'));

Convert the data set to a groupedData object, which is the required data format for the fitting function sbiofit. A groupedData object allows you to set independent variable and group variable names (if they exist). Set the units of the ID, Time, CentralConc, PeripheralConc, Age, and Sex variables. The units are optional and only required for the UnitConversion feature, which automatically converts matching physical quantities to one consistent unit system.

gData = groupedData(data);
gData.Properties.VariableUnits = {'','hour','milligram/liter','milligram/liter','',''};

The IndependentVariableName and GroupVariableName properties have been automatically set to the Time and ID variables of the data.

gData.Properties
ans = 

                Description: ''
       VariableDescriptions: {}
              VariableUnits: {''  'hour'  'milligram/liter'  'milligram/liter'  ''  ''}
             DimensionNames: {'Row'  'Variable'}
                   UserData: []
                   RowNames: {}
              VariableNames: {'ID'  'Time'  'CentralConc'  'PeripheralConc'  'Sex'  'Age'}
          GroupVariableName: 'ID'
    IndependentVariableName: 'Time'

For illustration purposes, use the first five individual data for training and the 6th individual data for testing.

trainData = gData([gData.ID < 6],:);
testData  = gData([gData.ID == 6],:);

Display the response data for each individual in the training set.

sbiotrellis(trainData,'ID','Time',{'CentralConc','PeripheralConc'});

Use the built-in PK library to construct a two-compartment model with infusion dosing and first-order elimination where the elimination rate depends on the clearance and volume of the central compartment. Use the configset object to turn on unit conversion.

pkmd                                    = PKModelDesign;
pkc1                                    = addCompartment(pkmd,'Central');
pkc1.DosingType                         = 'Bolus';
pkc1.EliminationType                    = 'linear-clearance';
pkc1.HasResponseVariable                = true;
pkc2                                    = addCompartment(pkmd,'Peripheral');
model                                   = construct(pkmd);
configset                               = getconfigset(model);
configset.CompileOptions.UnitConversion = true;

Assume every individual receives a bolus dose of 100 mg at time = 0.

dose             = sbiodose('dose','TargetName','Drug_Central');
dose.StartTime   = 0;
dose.Amount      = 100;
dose.AmountUnits = 'milligram';
dose.TimeUnits   = 'hour';

The data contains measured plasma concentration in the central and peripheral compartments. Map these variables to the appropriate model components, which are Drug_Central and Drug_Peripheral.

responseMap = {'Drug_Central = CentralConc','Drug_Peripheral = PeripheralConc'};

Specify the volumes of central and peripheral compartments Central and Peripheral, intercompartmental clearance Q12, and clearance Cl_Central as parameters to estimate. The estimatedInfo object lets you optionally specify parameter transforms, initial values, and parameter bounds. Since both Central and Peripheral are constrained to be positive, specify a log-transform for each parameter.

paramsToEstimate    = {'log(Central)', 'log(Peripheral)', 'Q12', 'Cl_Central'};
estimatedParam      = estimatedInfo(paramsToEstimate,'InitialValue',[1 1 1 1]);

Use the 'CategoryVariableName' property of the estimatedInfo object to specify which category to use during fitting. Use 'Sex' as the group to fit for the clearance Cl_Central and Q12 parameters. Use 'Age' as the group to fit for the Central and Peripheral parameters.

estimatedParam(1).CategoryVariableName = 'Age';
estimatedParam(2).CategoryVariableName = 'Age';
estimatedParam(3).CategoryVariableName = 'Sex';
estimatedParam(4).CategoryVariableName = 'Sex';
categoryFit = sbiofit(model,trainData,responseMap,estimatedParam,dose)

When fitting by category (or group), sbiofit always returns one results object, not one for each category level. This is because both male and female individuals are considered to be part of the same optimization using the same error model and error parameters, similarly for the young and old individuals.

categoryFit = 

  OptimResults with properties:

                   ExitFlag: 3
                     Output: [1x1 struct]
                  GroupName: []
                       Beta: [8x5 table]
         ParameterEstimates: [20x6 table]
                          J: [40x8x2 double]
                       COVB: [8x8 double]
           CovarianceMatrix: [8x8 double]
                          R: [40x2 double]
                        MSE: 0.1240
                        SSE: 8.9269
                    Weights: []
    EstimatedParameterNames: {'Central'  'Peripheral'  'Q12'  'Cl_Central'}
             ErrorModelInfo: [1x3 table]
         EstimationFunction: 'lsqnonlin'
                 ErrorModel: 'constant'
            ErrorParameters: [1x1 table]

Plot the category-specific estimated results. For the Cl_Central and Q12 parameters, all males had the same estimates, and similarly for the females. For the Central and Peripheral parameters, all young individuals had the same estimates, and similarly for the old individuals.

plot(categoryFit);

As for testing purposes, simulate the responses of the 6th individual who is an old male. Since you have estimated one set of parameters for the Age category (young versus old), and another set for Sex category (male versus female), you can simulate the responses of an old male even though there is no such individual in the training data.

Use the predict method to simulate the responses. ynew contains simulation data and paramestim contains parameter estimates used during simulation.

[ynew,paramestim] = predict(categoryFit,testData,dose);

Plot the simulated responses of the old male.

sbioplot(ynew);

The paramestim variable contains the estimated parameters used by the predict method. The parameter estimates for corresponding categories were obtained from the categoryFit.ParameterEstimates property. Specifically, Central and Peripheral parameter estimates are obtained from the Old group, and Q12 and Cl_Central parameter estimates are obtained from the Male group.

paramestim
paramestim = 

        Name        Estimate    StandardError    Group    CategoryVariableName    CategoryValue
    ____________    ________    _____________    _____    ____________________    _____________

    'Central'        1.1993     0.0050591        6        'Age'                   Old          
    'Peripheral'     0.5627       0.02908        6        'Age'                   Old          
    'Q12'            1.3119      0.062733        6        'Sex'                   Male         
    'Cl_Central'    0.56631     0.0079784        6        'Sex'                   Male  

Overlay the experimental results on the simulated data.

figure;
plot(testData.Time,testData.CentralConc,'LineStyle','none','Marker','d','MarkerEdgeColor','b');
hold on
plot(testData.Time,testData.PeripheralConc,'LineStyle','none','Marker','d','MarkerEdgeColor','r');
plot(ynew.Time,ynew.Data(:,1),'b');
plot(ynew.Time,ynew.Data(:,2),'r');
hold off
legend({'OBS1(CentralConc)','OBS2(PeripheralConc)',...
        'PRED1(Central.Drug\_Central)','PRED2(Peripheral.Drug\_Peripheral)'});

References

[1] Yi, T-M., Kitano, H., and Simon, M. (2003). A quantitative characterization of the yeast heterotrimeric G protein cycle. PNAS. 100, 10764–10769.

Introduced in R2014a