search

Open-Systems-Pharmacology / Forum

Discussion forum for the Open Systems Pharmacology Project
65 stars 19 forks source link

Can't run model with Lipophilicity value > 10 #696

Closed jonathanlsn closed 3 years ago

jonathanlsn commented 3 years ago

Hello,

I recently met an issue with the logP I have to set in compound parameters on PK-sim. I wanted to set this value to 11.2 but the software return this error "Value for lipophilicity should be less than or equal to 10 Log Units". I would like to know why a value higher than 10 can't be handled by the software ?

Thank you in advance

tobiasK2001 commented 3 years ago

Dear Jonathan,

why do you think your compound could hav a logP of 11.2? PK-Sim sets a limit here because a logP > 10 would be well outside the accepted possible ADMET space of NCEs. See here for more details https://pubmed.ncbi.nlm.nih.gov/11259830/

best, Tobias

StephanSchaller commented 3 years ago

@jonathanlsn, extending the comment ba @tobiasK2001, the reason for PK-Sim to limit this range is related to "model confidence". The partitioning methods contained in PK-Sim are developed for pharmaceutical NCEs where the "Rule of 5" referred to by @tobiasK2001 are mostly adhered to (LogP < 5 is one of these rules). This means that the PK-Sim predicted partitioning based on a LogP>5 might be wrong, as the partition coefficient methods available in PK-Sim are not "qualified" for such compounds/LogP ranges. We have worked with a couple of compounds (organic chemicals) with LogP > 5 (but < 10) and looked into other partitioning methods (e.g. Poulin & Haddad) to better characterize the ADME, but there are quite a few additional aspects that make things complicated for highly lipophilic chemicals.