Quantitative Structure-Activity Relationship

Structure-based drug design is becoming an essential tool for faster and more cost-efficient lead discovery relative to the traditional method. Genomic, proteomic, and structural studies have provided hundreds of new targets and opportunities for future drug discovery. This situation poses a major problem: the necessity to handle the "big data" generated by combinatorial chemistry. Artificial intelligence (AI) and deep learning play a pivotal role in the analysis and systemization of larger data sets by statistical machine learning methods. Advanced AI-based sophisticated machine learning tools have a significant impact on the drug discovery process including medicinal chemistry. In this review, we focus on the currently available methods and algorithms for structure-based drug design including virtual screening and de novo drug design, with a special emphasis on AI- and deep-learning-based methods used for drug discovery.

BackgroundQuantitative structure-activity relationship (QSAR) is a computational modeling method for revealing relationships between structural properties of chemical compounds and biological activities. QSAR modeling is essential for drug discovery, but it has many constraints. Ensemble-based machine learning approaches have been used to overcome constraints and obtain reliable predictions. Ensemble learning builds a set of diversified models and combines them. However, the most prevalent approach random forest and other ensemble approaches in QSAR prediction limit their model diversity to a single subject.ResultsThe proposed ensemble method consistently outperformed thirteen individual models on 19 bioassay datasets and demonstrated superiority over other ensemble approaches that are limited to a single subject. The comprehensive ensemble method is publicly available at http://data.snu.ac.kr/QSAR/ .ConclusionsWe propose a comprehensive ensemble method that builds multi-subject diversified models and combines them through second-level meta-learning. In addition, we propose an end-to-end neural network-based individual classifier that can automatically extract sequential features from a simplified molecular-input line-entry system (SMILES). The proposed individual models did not show impressive results as a single model, but it was considered the most important predictor when combined, according to the interpretation of the meta-learning.

The present article reviews published efforts to study acetylcholinesterase and butyrylcholinesterase structure and function using computer-based modeling and simulation techniques. Structures and models of both enzymes from various organisms, including rays, mice, and humans, are discussed to highlight key structural similarities in the active site gorges of the two enzymes, such as flexibility, binding site location, and function, as well as differences, such as gorge volume and binding site residue composition. Catalytic studies are also described, with an emphasis on the mechanism of acetylcholine hydrolysis by each enzyme and novel mutants that increase catalytic efficiency. The inhibitory activities of myriad compounds have been computationally assessed, primarily through Monte Carlo-based docking calculations and molecular dynamics simulations. Pharmaceutical compounds examined herein include FDA-approved therapeutics and their derivatives, as well as several other prescription drug derivatives. Cholinesterase interactions with both narcotics and organophosphate compounds are discussed, with the latter focusing primarily on molecular recognition studies of potential therapeutic value and on improving our understanding of the reactivation of cholinesterases that are bound to toxins. This review also explores the inhibitory properties of several other organic and biological moieties, as well as advancements in virtual screening methodologies with respect to these enzymes.

Imidazoles have been proven to be very effective pesticides, especially against phytopathogenic fungi and insects. Due to their negative effects on the environment, only a few imidazoles have been approved for use by the European Commission (EC). There is an urgent need to develop new imidazole derivatives with high efficiency and a wide spectrum of action against numerous pests that are, at the same time, safe for the environment and beneficial for organisms and humans. In order to reduce expensive and time-consuming experiments, an in silico approach based on quantitative structure-activity relationship (QSAR) models is valuable for predicting the toxicity of new or untested chemicals. In this study, we used the Vega and ChemFREE web platforms to evaluate the pesticide similarity, environmental risk properties, and ecotoxicological effects of imidazole derivatives designed for potential synthesis. Adamantane-, alkyl-, and triazole-amide, ester, carbamate, and ketone derivatives were filtered for the evaluated properties, and four alkyl-amides were highlighted as potentially effective and environmentally safe antifungal, herbicidal, and insecticidal agents. Molecular docking studies indicated the possible mechanism of action of the antifungal, herbicidal, insecticidal, and antibacterial activities of the observed compounds and revealed structural features important for binding to specific receptors.