Fragment contribution models for predicting skin permeability using HuskinDB.

Area of Science:

• Pharmacokinetics
• Computational Chemistry
• Dermatology

Background:

• Traditional skin permeation models rely on animal data and physicochemical properties.
• Predicting human skin permeability is crucial for drug development and chemical safety assessment.
• Existing methods can be complex and may not fully represent human physiology.

Purpose of the Study:

• To develop a novel, non-animal model for predicting human skin permeability.
• To utilize a freely accessible human skin permeation database (HuskinDB).
• To establish a simplified prediction method based on chemical fragment data.

Main Methods:

• Developed fragment-based models using functional group data from the HuskinDB database.
• Analyzed common functional groups and their calculable effects on skin permeability.
• Validated models using compounds with known human skin permeation values.

Main Results:

• Created predictive models based solely on the type and number of functional groups.
• Demonstrated that physicochemical properties are not required for prediction with these models.
• Achieved accurate predictions for human skin permeability (logKp) for various compounds.

Conclusions:

• Fragment-based models offer a simplified and non-animal approach to predicting human skin permeability.
• This method reduces reliance on animal testing and complex physicochemical property calculations.
• The developed models are applicable to compounds containing analyzed functional groups, streamlining permeability assessment.