Role of physicochemical properties in the estimation of skin permeability:in vitrodata assessment by Partial Least-Squares Regression

Abstract

Skin provides passage for the delivery of drugs. The in vitro and in vivo testing of chemicals for estimation of dermal absorption is very time consuming, costly and has many ethical difficulties related to human and animal testing. The solution to the problem is Quantitative structure-permeability relationships. This method relates dermal penetration properties of a range of chemical compounds to their physicochemical parameters. In the present study, an effort has been made to develop models for the accurate prediction of skin permeability using a large, diverse dataset through the combination of various regression methods coupled with the Genetic Algorithm (GA)/Interval Partial Least-Squares Algorithm (iPLS). The descriptors were calculated using e-DRAGON and ADME Pharma Algorithms-Abrahams descriptors. The original dataset was divided into a training set and a testing set using the Kennard-Stone Algorithm. The selection of descriptors was made by the GA and iPLS. The model applicability domain was determined. The results showed that a three-parameter model built through Partial Least-squares Regression was most accurate with r(2) of 0.936.

Data Quality in the Human and Environmental Health Sciences: Using Statistical Confidence Scoring to Improve QSAR/QSPR Modeling

“…14,15 More human in vitro k p data have inevitably become available in the two and half decades since Flynn's seminal publication 14,[16][17][18] , thus the QSPR can be reassessed and rebuilt with a greater consideration and understanding of data quality.…”

“…14,[16][17][18] All k p values were converted to a standard unit (cm/h). Duplicate log k p values (and those within ± 0.01 cm/h) were removed as they are most likely to be derived from the same source.…”

Predicting chemically-induced skin reactions. Part II: QSAR models of skin permeability and the relationships between skin permeability and skin sensitization

“…In vitro human skin permeability coefficients were retrieved from the literature ( Chauhan and Shakya, 2010 ) including 211 records expressed in logKp (cm.h −1 ); this dataset contained the well-known and frequently studied Flynn dataset ( Flynn, 1990 ). 17 duplicates and two sets of triplicates were identified and curated leaving unique compounds only.…”

“…Another recent study showed that most of the available QSAR models underestimate the skin permeability of hydrophilic solutes ( Chen et al, 2013 ). Recent studies ( Chauhan and Shakya, 2010 ) attempted to overcome the related problems with complex chemical mixtures and built a QSAR model based on several mixtures of 36 chemicals with porcine skin data. The latter model followed the best practices of QSAR modeling ( Tropsha, 2010 ); however, we have identified 21 duplicative structures in that dataset (see Table S3 ) indicating a potential bias of the model and a likely over-estimation of its true performance.…”

Modeling and Prediction of Solvent Effect on Human Skin Permeability using Support Vector Regression and Random Forest

“…During the early developmental stages of dermatological medicines or cosmetics, when numerous candidate compounds need to be evaluated, efficiently estimating the skin permeability of these compounds is particularly important. Therefore, various prediction models of skin permeability have been established (7)(8)(9)(10)(11), which provide efficient and accurate alternatives to animal experiments. Among the most famous classical models is the Potts and Guy's model (11), which predicts the permeability coefficient (k p ) by linear regression using octanol-water partition coefficients (log P) and molecular weights of permeants.…”