Simutaneous parameter estimation for multiple compartments and multiple dataset

1 view (last 30 days)
Nan Wu
Nan Wu on 17 Jun 2020
Answered: Arthur Goldsipe on 19 Jun 2020
I am working on PK/PD analysis for a parent drug with its metabolites. Therefore I have 3 observations along with time changes, which are the concentration of parent drug, metabolite concentration and target activity. In addition, I have 3 datasets which contains the time and mean values of those above 3 observations. The 3 datasets were seperated by different dose levels.
Now I am trying to use nlinfit to do parameter estimation, as well as to get confidence intervals of each parameters. Here are some code I have written. However, I realized that it cannot neither do simutaneous fitting for concentration of parent drug, metabolite concentration and target activity or for different datasets.
[beta, resid, J]=nlinfit(tObs,aObs, @model, [1 2], options);
I find some instructions about sbiofit which could pool data, would it be helpful for different dataset analysis? However, I still have no any clue about simutanoues fitting for multiple comparments (concentration of parent drug, metabolite concentration and target activity).
I would rather write code manuly, but not use simbiology GUI. Anyone have some experience on this? Any suggestion would be appreciated!

Answers (1)

Arthur Goldsipe
Arthur Goldsipe on 19 Jun 2020
Depending on the details of the estimation, you might be able to use sbiofit for this sort of problem by merging the 3 datasets into one dataset. sbiofit allows you to specify different dosing conditions and to estimate parameters that apply to all conditions as well as estimating parameters that are different for various experimental conditions. The main limitation is that you would not be able to prescibe parameter values that are different for the different experimental conditions. (For example, you could not prescribe that a particular parameter value was 1.0 for condition A, 2.0 for condition B, and 3.0 for condition C, but you could estimate different parameter values for each of these conditions.)
Can you take a look at the documentation and the examples for sbiofit? If you still have questions on how to use it, I suggest creating a new MATLAB Answers question with more details on exactly what you're trying to do. If you can provide an example problem (data, model, etc.), then it will be much easier for other people to help you.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!