The application of Gaussian processes in the prediction of percutaneous absorption

Abstract

A non-linear approach was more appropriate than QSPRs or SLNs for the analysis of the dataset employed herein, as the prediction and confidence values in the prediction given by the Gaussian process are better than with other methods examined. Gaussian process provides a novel way of analysing skin absorption data that is substantially more accurate, statistically robust and reflective of our empirical understanding of skin absorption than the QSPR methods so far applied to skin absorption.

Spontaneous chemical degradation of substance P in the solid phase and in solution

“…Acetate and other weak carboxylic acids were found to catalyze the diketopiperazine formation of polymer-bound dipeptide esters (25, 26) in solidphase synthesis, whereas hydrochloride and trifluoroacetate salts were found to be considerably more stable (25). However, acetate buffer was much less effective than phosphate buffer at the same pH in catalyzing the cyclization of His-Pro-NHz (23). In agreement with these results, the transformation of lyophilized SP acetate samples into the lyophilized hydrochlorides or trifluoroacetates increased the stability of the undecapeptide dramatically.…”

“…In a few cases the spontaneous cyclization may also play a role in the biosynthetic pathway of naturally occurring diketopiperazines. Thus cyclo(His-Pro), a biologically active cyclic dipeptide detected in a variety of tissues and body fluids of animals and man, is thought to be derived nonenzymatically from His-Pro-NHz, which is formed through the action of pyroglutamate aminopeptidase on TRH, < Glu-His-Pro-NHz (23).…”

Predicting Skin Permeability by Means of Computational Approaches: Reliability and Caveats in Pharmaceutical Studies

“…Machine learning methods can include the gaussian process regression with automatic resonance detection (GPRARD). This method was compared in different studies 143,144 to gaussian process regression (GPR), and single linear networks (SLN). The use of GPR, in particular GPRARD, is able to quantify the covariance and the length scale of each descriptor in the model, giving a deeper comprehension of the significance of each feature 144 .…”

“…This method was compared in different studies 143,144 to gaussian process regression (GPR), and single linear networks (SLN). The use of GPR, in particular GPRARD, is able to quantify the covariance and the length scale of each descriptor in the model, giving a deeper comprehension of the significance of each feature 144 . GPR methods have been applied to the exploration of skin membranes datasets other than human skin, and to investigate how the nature of the dataset may influence its analysis 145 .…”

The Contributions of the Celtic Masters and Their Associates

“…More recently Pugh's work has focused on machine learning methods of analysis, with an emphasis on both modelling percutaneous absorption with global models [83] and the use of support vector machines and other machine learning methods to classify and predict the benefits of enhancers of percutaneous absorption [84]. …”