Computer-aided formulation design is a methodology that utilizes domain knowledge and selected methods and tools suitable for computer-based applications to assist in formulation (product) design. In this paper, molecular dynamics simulation and Bayesian neural network algorithms are combined with well-known engineering models to help accelerate the development and optimization of formulation-based detergent products with a view to improve product quality and performance. In particular, the mechanism of the behavior of polymers (an active ingredient in the product) to improve the product quality in terms of the fragrance and its residence time is highlighted. Results from molecular dynamic simulation applied to study the molecular interaction mechanism show that the polymers have an attraction effect with fragrance molecules and could adsorb more to make them to stay on the surface of clothes. In addition, the polymer attenuates the diffusion of the fragrance molecules, lengthening the entire process of fragrance diffusion, which is the essence of the ability of the polymer to slow down the release of the fragrance. A Quantitative Structure-Property Relationship (QSPR) model between component proportions and fragrance diffusion is established through Bayesian Neural Network (BNN) and the product formulation is optimized based on this model. Keeping polymer and perfume ingredients unchanged, the surfactant amounts are optimized to provide improved product quality.

Structure-based virtual screening of binding of candidate drug molecules is a topic of increasing interest in the discovery of small molecule drugs. As the same drug molecule may bind to different binding sites on a target protein, the binding site selectivity that is related to the binding tendency of candidate drug molecules to different binding sites after reaching the target protein need to be considered in sufficient details. In this work, a systematic and computer-aided virtual screening framework based on the binding site selectivity to screen candidate drug molecules in terms of their ability to bind on selected sites is presented. The framework integrates two machine learning (ML)-based models to predict the binding potential and binding selectivity to specific binding sites that are important for virtual screening of drug molecules. The details of the ML-based models together with the work-flow of the computer-aided virtual screening methods and the efficient and consistent integration of related drug design tools are presented. The applicability of this virtual screening framework is illustrated through a case study involving the screening for drug molecules as inhibitors to block the binding of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) to angiotensin converting enzyme 2 (ACE2), which is the target protein. The case study results point to identification of new candidate inhibitors with better binding site selectivity than two known potential inhibitors, Nilotinib and SSAA09E2.