Use of three-dimensional descriptors in molecular design for biologically active compounds

Prediction method of pharmacokinetic parameters of small molecule drugs based on GCN network model

CONTEXT

Accurately predicting plasma protein binding rate (PPBR) and oral bioavailability (OBA) helps to better reveal the absorption and distribution of drugs in the human body and subsequent drug design. Although machine learning models have achieved good results in prediction accuracy, they often suffer from insufficient accuracy when dealing with data with irregular topological structures.

METHODS

In view of this, this study proposes a pharmacokinetic parameter prediction framework based on graph convolutional networks (GCN), which predicts the PPBR and OBA of small molecule drugs. In the framework, GCN is first used to extract spatial feature information on the topological structure of drug molecules, in order to better learn node features and association information between nodes. Then, based on the principle of drug similarity, this study calculates the similarity between small molecule drugs, selects different thresholds to construct datasets, and establishes a prediction model centered on the GCN algorithm. The experimental results show that compared with traditional machine learning prediction models, the prediction model constructed based on the GCN method performs best on PPBR and OBA datasets with an inter-molecular similarity threshold of 0.25, with MAE of 0.155 and 0.167, respectively. In addition, in order to further improve the accuracy of the prediction model, GCN is combined with other algorithms. Compared to using a single GCN method, the distribution of the predicted values obtained by the combined model is highly consistent with the true values. In summary, this work provides a new method for improving the rate of early drug screening in the future.

In-Silico Approaches for the Screening and Discovery of Broad-Spectrum Marine Natural Product Antiviral Agents Against Coronaviruses

The urgent need for SARS-CoV-2 controls has led to a reassessment of approaches to identify and develop natural product inhibitors of zoonotic, highly virulent, and rapidly emerging viruses. There are yet no clinically approved broad-spectrum antivirals available for beta-coronaviruses. Discovery pipelines for pan-virus medications against a broad range of betacoronaviruses are therefore a priority. A variety of marine natural product (MNP) small molecules have shown inhibitory activity against viral species. Access to large data caches of small molecule structural information is vital to finding new pharmaceuticals. Increasingly, molecular docking simulations are being used to narrow the space of possibilities and generate drug leads. Combining in-silico methods, augmented by metaheuristic optimization and machine learning (ML) allows the generation of hits from within a virtual MNP library to narrow screens for novel targets against coronaviruses. In this review article, we explore current insights and techniques that can be leveraged to generate broad-spectrum antivirals against betacoronaviruses using in-silico optimization and ML. ML approaches are capable of simultaneously evaluating different features for predicting inhibitory activity. Many also provide a semi-quantitative measure of feature relevance and can guide in selecting a subset of features relevant for inhibition of SARS-CoV-2.

Molecular Descriptors as a Facile Tool toward Designing Surface-Functionalized Nanoparticles for Drug Delivery

Modulating the surface chemistry of nanoparticles, often by grafting hydrophilic polymer brushes (e.g., polyethylene glycol) to prepare nanoformulations that can resist opsonization in a hematic environment and negotiate with the mucus barrier, is a popular strategy toward developing biocompatible and effective nano-drug delivery systems. However, there is a need for tools that can screen multiple surface ligands and cluster them based on both structural similarity and physicochemical attributes. Molecular descriptors offer numerical readouts based on molecular properties and provide a fertile ground for developing quick screening platforms. Thus, a study was conducted with 14 monomers/repeating blocks of polymeric chains, namely, oxazoline, acrylamide, vinylpyrrolidone, glycerol, acryloyl morpholine, dimethyl acrylamide, hydroxypropyl methacrylamide, hydroxyethyl methacrylamide, sialic acid, carboxybetaine acrylamide, carboxybetaine methacrylate, sulfobetaine methacrylate, methacryloyloxyethyl phosphorylcholine, and vinyl-pyridinio propanesulfonate, capable of imparting hydrophilicity to a surface when assembled as polymeric brushes. Employing free, Web-based, and user-friendly platforms, such as SwissADME and ChemMine tools, a series of molecular descriptors and Tanimoto coefficient of molecular pairs were determined, followed by hierarchical clustering analyses. Molecular pairs of oxazoline/dimethyl acrylamide, hydroxypropyl methacrylamide/hydroxyethyl methacrylamide, acrylamide/glycerol, carboxybetaine acrylamide/vinyl-pyridinio propanesulfonate, and sulfobetaine methacrylate/methacryloyloxyethyl phosphorylcholine were clustered together. Similarly, the molecular pair of hydroxypropyl methacrylamide/hydroxyethyl methacrylamide demonstrated a high Tanimoto coefficient of >0.9, whereas the pairs oxazoline/vinylpyrrolidone, acrylamide/dimethyl acrylamide, acryloyl morpholine/dimethyl acrylamide, acryloyl morpholine/hydroxypropyl methacrylamide, acryloyl morpholine/hydroxyethyl methacrylamide, carboxybetaine methacrylate/sulfobetaine methacrylate, and glycerol/hydroxypropyl methacrylamide had a Tanimoto coefficient of >0.8. The analyzed data not only demonstrated the ability of such in silico tools as a facile technique in clustering molecules of interest based on their structure and physicochemical characteristics but also provided vital information on their behavior within biological systems, including the ability to engage an array of possible molecular targets when the monomers are self-assembled on nanoparticulate surfaces.

A machine learning regression model for the screening and design of potential SARS-CoV-2 protease inhibitors

The widespread infection caused by the 2019 novel corona virus (SARS-CoV-2) has initiated global efforts to search for antiviral agents. Drug discovery is the first step in the development of commercially viable pharmaceutical products to deal with novel diseases. In an effort to accelerate the screening and drug discovery workflow for potential SARS-CoV-2 protease inhibitors, a machine learning model that can predict the binding free energies of compounds to the SARS-CoV-2 main protease is presented. The optimized multiple linear regression model, which was trained and tested on 226 natural compounds demonstrates reliable prediction performance (r 2 test = 0.81, RMSE test = 0.43), while only requiring five topological descriptors. The externally validated model can help conserve and maximize available resources by limiting biological assays to compounds that yielded favorable outcomes from the model. The emergence of highly infectious diseases will always be a threat to human health and development, which is why the development of computational tools for rapid response is very important.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s13721-021-00326-2.