simbiology parameter estimation with dosing schedule

Hi, I have some PK data for that I'd like to fit to a 2-comp model. The data includes an initial dose at t=0 and another (different) dose at t>0. How can I use the paramfit task in simbiology with a dosing schedule? I can't see the option and it is trying to fit the concentration curves with a single initial dose which is obviously poor.

Thanks!

[Arthur Goldsipe]

Can you clarify a few things? What version of MATLAB are you using? And are you using the SimBiology Desktop or the command line? Does your data include different "groups" (different patients or different experiments)? If so, do you want to estimate the parameters for each individual separately, one set of parameters for all individuals, or some sort of mixed effects model (with population-average parameters and individual-specific "random" effects)? You can do all of these things in SimBiology, but you first need to decide which of those you want to do.

Answer 1

[Arthur Goldsipe]

Let me go ahead and take a stab at answering this. With additional clarifications, I can offer a more tailored example.

You usually have limited experimental data on each individual in your PK data set, so you typically use a non-linear mixed effects model to simultaneously fit all patients to your 2-compartment model. In R2013a of SimBiology, here's a quick outline of how to do this:

1. Create a 2-compartment model by selecting "Create PK Model" under the "Add Model" button in the SimBiology Desktop's toolstrip.
2. Adjust the model as necessary (for example, by setting the compartment volumes to some reasonable value). However, do not set the Dose species' initial amount in the model, since SimBiology determines the dosing schedule from the experimental data.
3. Load your experimental data (which includes the doses at all times, including time=0 and time>0) by selecting "Load data from file" under the "Add Data" button in the toolstrip.
4. Ensure that the data is properly configured, marking the column with patient ID's as the group variable and the time column as the independent variable.
5. Add a "Fit data" task by selecting this menu item from the Add Task button in the toolstrip.
6. Configure the fit task. For example, switch the Estimation Method to "population fit (NLMEFIT)" if you need to do a mixed-effects model. Update initial parameter estimates as necessary. Ensure the correct species is the model is being dosed using the correct column from your data set. Ensure the correct species and the correct column from your data set are marked as the "response" (the concentrations you use for fitting). Change any other options.
7. Run the fit task. Change any options and re-run as necessary.

You may also find the following links helpful:

• PKPD modeling in SimBiology
• Performing PKPD fitting in SimBiology

[Sietse Braakman]

Please have a look at this entry on our Community website as well with guidance on parameter estimation in SimBiology

https://www.mathworks.com/matlabcentral/discussions/simbiology/134-guidance-and-troubleshooting-for-parameter-estimation-in-simbiology