8 replies to this topic

[j.pereira]

Dear All,

 

I have a model which works in Berkeley Madonna using the STIFF Solver with settings DTMIN = 10^-6 and DTMAX = 1. It is a TMDD model for a mAb which including an endossome compartment for Fc-Rn recycling and target binding both in the central compartment and endossome compartment. The model works fine and I get the following results using Berkeley Madonna.

 

 

I wrote the model in Phoenix and I get strange results:

 

 

I have already checked the Phoenix model code and the parameters thourougly, and I don't find a coding mistake. What also leads me to think that there is a problem with the ODE Solver is the fact that when chaging the options of the ODE Solver  in Phoenix (Set stiff and ODE max step to 1) I get even weirder results, for example:

 

 

Also, the model does not work properly with the standard ODE Solver in Berkeley Madonna and I have to change to stiff.

 

That leads me to think that the ODE Solver in Phoenix is not being able to solve this model. Or at least I don't know what else I can change in the Phoenix settings to solve this.

 

I would really like to get this model to work in Phoenix, so any help would be welcome!

 

Would it be ok for you if I send the Phoenix file to support@certara.com?

 

Thank you very much,

Joao

 

 

[Simon Davis]

Yes Joao, please send the file to support, if you do not want to post it here publically

 

Please mention in the email that you have started the conversation on here so we can ensure that we feed back any suggestions/findings to the other readers too.

 

  Simon

[j.pereira]

Hi Simon,

 

I have sent the model to support@certara. I hope you guys can find out where the problem is...

 

Best,

Joao

Yes Joao, please send the file to support, if you do not want to post it here publically

 

Please mention in the email that you have started the conversation on here so we can ensure that we feed back any suggestions/findings to the other readers too.

 

  Simon

[smouksassi1]

also please send the B-M code so we can double check that your model was recoded fine into pml.

Samer

[serge guzy]

Yes before assuming there is a problem with ODE, we need to be sure the code is the same. I tried your model with the population ODE, with matrix exponent, stiff, non stiff and all gave me the same answer.

Then I suspect we are talking about 2 different codes.

Best

Serge

[j.pereira]

Hi Serge and Samer,

 

I just sent you the BM code to support@certara.

 

Thank you very much.

 

Joao

[serge guzy]

Here is my translated code in Phoenix.

All seems good to me. i am getting the classical TMDD profile.

best Regards

Serge

Attached Files

•  jo_pereira.phxproj   348.21KB   375 downloads

[j.pereira]

Thanks you very much Serge! It also works fine with me, I changed your code by adding dosepoint() and I can also simulate multiple doses exactly as in Berkeley Madonna.

 

It is still not clear what was wrong with my first code, but it seems that it is not the programs fault

 

Best,

 

Joao

[serge guzy]

Yes indeed, it was not the program fault. You know, sometime it is waste of time trying to understand what you did wrong.

I started from scratch and it was successful. This is what is important.

Good luck

Serge