Molecular property prediction based on graph structure learning.

MOTIVATION: Molecular property prediction (MPP) is a fundamental but challenging task in the computer-aided drug discovery process. More and more recent works employ different graph-based models for MPP, which have achieved considerable progress in improving prediction performance. However, current models often ignore relationships between molecules, which could be also helpful for MPP. RESULTS: For this sake, in this article we propose a graph structure learning (GSL) based MPP approach, called GSL-MPP. Specifically, we first apply graph neural network (GNN) over molecular graphs to extract molecular representations. Then, with molecular fingerprints, we construct a molecule similarity graph (MSG). Following that, we conduct GSL on the MSG, i.e. molecule-level GSL, to get the final molecular embeddings, which are the results of fuzing both GNN encoded molecular representations and the relationships among molecules. That is, combining both intra-molecule and inter-molecule information. Finally, we use these molecular embeddings to perform MPP. Extensive experiments on 10 various benchmark datasets show that our method could achieve state-of-the-art performance in most cases, especially on classification tasks. Further visualization studies also demonstrate the good molecular representations of our method. AVAILABILITY AND IMPLEMENTATION: Source code is available at https://github.com/zby961104/GSL-MPP.

GPR91, a critical signaling mechanism in modulating pathophysiologic processes in chronic illnesses.

Succinate receptor GPR91 is one of G protein-coupled receptors (GPCRs), and is expressed in a variety of cell types and tissues. Succinate is its natural ligand, and its activation represents that an intrinsic metabolic intermediate exerts a regulatory role on many critical life processes involving pathophysiologic mechanisms, such as innate immunity, inflammation, tissue repair, and oncogenesis. With the illustration of 3-dimensional crystal structure of the receptor and discovery of its antagonists, it is possible to dissect the succinate-GPR91-G protein signaling pathways in different cell types under pathophysiological conditions. Deep understanding of the GPR91-ligand binding mode with various agonists and antagonists would aid in elucidating the molecular basis of a spectrum of chronic illnesses, such as hypertension, diabetes, and their renal and retina complications, metabolic-associated fatty liver diseases, such as nonalcoholic steatohepatitis and its fibrotic progression, inflammatory bowel diseases (Crohn's disease and ulcerative colitis), age-related macular degeneration, rheumatoid arthritis, and progressive behaviors of malignancies. With better delineation of critical regulatory role of the succinate-GPR91 axis in these illnesses, therapeutic intervention may be developed by specifically targeting this signaling pathway with small molecular antagonists or other strategies.

The Use of Computational Approaches in the Discovery and Mechanism Study of Opioid Analgesics.

Opioid receptors that belong to class A G protein-coupled receptors (GPCRs) are vital in pain control. In the past few years, published high-resolution crystal structures of opioid receptor laid a solid basis for both experimental and computational studies. Computer-aided drug design (CADD) has been established as a powerful tool for discovering novel lead compounds and for understanding activation mechanism of target receptors. Herein, we reviewed the computational-guided studies on opioid receptors for the discovery of new analgesics, the structural basis of receptor subtype selectivity, agonist interaction mechanism, and biased signaling mechanism.

Discovery of aryl-piperidine derivatives as potential antipsychotic agents using molecular hybridization strategy.

Schizophrenia is a chronic, disabling mental disorder that affects about one percent of world's population. Drugs acting on multiple targets have been demonstrated to provide superior efficacy in schizophrenia than agents acting on single target. In this study, based on FW01, a selective potent 5-HT1A receptor agonist discovered via dynamic pharmacophore-based virtual screening, molecular hybridization strategy was employed to optimize its in vitro activity over D2 and 5-HT2A receptors. The optimized compound 9f was found to show dual potent D2 and 5-HT2A receptors antagonistic activity. In addition, compound 9f showed good in vivo metabolic stability with t1/2 of 2 h in ICR mice and good capability to penetrate the blood-brain barrier with Kp value of 4.03. These results demonstrated that the dual D2 and 5-HT1A receptor antagonist 9f could serve as a promising lead compound to discover potent antipsychotic agents.