Perspective on a chemistry classification system for AI-assisted formulation development.

Journal: Journal of controlled release : official journal of the Controlled Release Society
Published Date:

Abstract

This perspective article draws a distinction between some of the well-known drug classification systems and a "Chemistry Classification System" (CCS). Rather than have drug classification based on some simple properties like solubility and permeability or route of systemic elimination, a CCS results in more than four or five classes and each class has distinct properties that impact formulation development. This perspective provides and outline of 13 classes, but a CCS is a flexible system that introduces a thought process for classification. The number of classes is not rigid, and chemists are encouraged to adapt these methods to their own situations. A CCS utilizes machine-learning models and artificial intelligence (AI) to estimate physicochemical properties that result in unique, frequently observed dissolution, Absorption, Distribution, Metabolism, and Excretion (ADME) properties to guide formulation development.

Authors

  • Michael B Bolger
    Simulations Plus, Inc., 42505 10(th) Street West, Lancaster, CA 93534, USA. Electronic address: michael.bolger@simulations-plus.com.