After export from PK-Sim, some "Property" parameters are in the PSV-BB

[PavelBal]

After exporting a PK-Sim model with expressed proteins, following parameters that are "Property" are created in the PSV-BB:

• Fraction expressed in blood cells membrane
• Fraction expressed on tissue-side membrane of vascular endothelium

What for? @msevestre It is a bug, don't you think?

[msevestre]

What's the logic again? We only create value in PSV for what parameters?

[PavelBal]

Global and local, if those are not defined by a formula. I actually know why this is explicitely added - in the molecules-BB the value of Fraction expressed in blood cells membrane is NaN and nans are now filtered out from PSV-BB.

The new expression logic makes it very hard to extend spatial structure - the relevent parameters

Fraction expressed interstitial
Fraction expressed intracellular
Relative expression

are not extended to new compartments. I understand WHY it has been implemented like this but I think we need another solution. We really need a way to define MoleculeProperties on organ level (i.e. not only for physical containers, but also for logical). In this case, the above mentioned parameters could be defined as "Organ"-properties. Let's maybe have a chat about it some time.

[StephanSchaller]

Could be solved by tags / container criteria as in #2112# ?

[msevestre]

Let's discuss this tomorrow during the meeting

[msevestre]

This will be resolved with usage of ContainerCriteria for Parameters in Molecule

[msevestre]

I don't understand this bug here. What is the expected behavior? Property parameter should not be added to the PSV correct?

[msevestre]

Fraction expressed in blood cells membrane is NaN just because it's a formula and the tag are defined with MOLECULE instead of ..

[msevestre]

oh I see. They should not be created because they are property

[RobinM92]

Hello,
I was wondering if this issue is the cause of unwanted behavior of relative expression when sending a simulation from pksim to mobi. E.g. in Open-Systems-Pharmacology/Forum#1151 I sent asimulation using relative expression of EGFR to MoBi, and all the relative expression values are attributed to intracellular expression in the tissues (probably due how it is coded connected to the interstitial expression). For now, the only work-around I have found is adding all relative expression values per tissue (in ISF, plasma and blood cells) manually, which is extremely cumbersome and errorprone. Of course this is relevant as we use rel. expression in the turnover reactions of receptors per tissue, if not they all converge to the same concentration...