GcForest-based compound-protein interaction prediction model and its application in discovering small-molecule drugs targeting CD47.

Identifying compound-protein interaction plays a vital role in drug discovery. Artificial intelligence (AI), especially machine learning (ML) and deep learning (DL) algorithms, are playing increasingly important roles in compound-protein interaction (CPI) prediction. However, ML relies on learning from large sample data. And the CPI for specific target often has a small amount of data available. To overcome the dilemma, we propose a virtual screening model, in which word2vec is used as an embedding tool to generate low-dimensional vectors of SMILES of compounds and amino acid sequences of proteins, and the modified multi-grained cascade forest based gcForest is used as the classifier. This proposed method is capable of constructing a model from raw data, adjusting model complexity according to the scale of datasets, especially for small scale datasets, and is robust with few hyper-parameters and without over-fitting. We found that the proposed model is superior to other CPI prediction models and performs well on the constructed challenging dataset. We finally predicted 2 new inhibitors for clusters of differentiation 47(CD47) which has few known inhibitors. The IC50s of enzyme activities of these 2 new small molecular inhibitors targeting CD47-SIRPα interaction are 3.57 and 4.79 µM respectively. These results fully demonstrate the competence of this concise but efficient tool for CPI prediction.

The Therapeutic Potential of Antioxidants in Chemotherapy-Induced Peripheral Neuropathy: Evidence from Preclinical and Clinical Studies.

As cancer therapies advance and patient survival improves, there has been growing concern about the long-term adverse effects that patients may experience following treatment, and concerns have been raised about such persistent, progressive, and often irreversible adverse effects. Chemotherapy is a potentially life-extending treatment, and chemotherapy-induced peripheral neuropathy (CIPN) is one of its most common long-term toxicities. At present, strategies for the prevention and treatment of CIPN are still an open problem faced by medicine, and there has been a large amount of previous evidence that oxidative damage is involved in the process of CIPN. In this review, we focus on the lines of defense involving antioxidants that exert the effect of inhibiting CIPN. We also provide an update on the targets and clinical prospects of different antioxidants (melatonin, N-acetylcysteine, vitamins, α-lipoic acid, mineral elements, phytochemicals, nutritional antioxidants, cytoprotectants and synthetic compounds) in the treatment of CIPN with the help of preclinical and clinical studies, emphasizing the great potential of antioxidants as adjuvant strategies to mitigate CIPN.

Targeting the TGF-ß signaling pathway for fibrosis therapy: a patent review (2015-2020).

INTRODUCTION: Fibrosis is a serious disease that occurs in many organs, such as kidney, liver and lung. The deterioration of these organs ultimately leads to death. Due to the complex mechanisms of fibrosis, research and development of antifibrotic drugs is difficult. One solution is to focus on core pathways, one of which is the TGF-ß signaling pathway. In virtually every type of fibrosis, TGF-ß signaling is recognized as a critical pathway. AREA COVERED: This review discusses patents on active molecules related to the TGF-ß signaling. Molecules targeting components related to the activation of TGF-ß are introduced. Several strategies preventing signal propagation from active TGF-ß to downstream targets are also introduced, including TGF-ß antibodies, TGF-ß ligand traps, and inhibitors of TGF-ß receptor kinases. Finally, molecules affecting downstream targets in both canonical and noncanonical TGF-ß signaling pathways are described. EXPERT OPINION: Since the approval of pirfenidone, targeting TGF-ß signaling has been anticipated as an effective therapy for fibrosis. The potential of this therapy has been further supported by emerging patents on the TGF-ß signaling. This pathway can be entirely inhibited, from the activation of TGF-ß to downstream signaling. Inhibiting TGF-ß signaling is expected to provide more effective treatments for fibrosis.

Discovery and evaluation of new compounds targeting ribosomal protein S1 in antibiotic-resistant Mycobacterium Tuberculosis.

The emergence of antibiotic-resistant Mycobacterium Tuberculosis (Mtb) infections compels new treatment strategies, of which targeting trans-translation is promising. During the trans-translation process, the ribosomal protein S1 (RpsA) plays a key role, and the Ala438 mutant is related to pyrazinamide (PZA) resistance, which shows its effects after being hydrolysed to pyrazinoic acid (POA). In this study, based on the structure of the RpsA C-terminal domain (RpsA-CTD) and POA complex, new compounds were designed. After being synthesized, the compounds were tested in vitro with saturation transfer difference (STD), fluorescence quenching titration (FQT) and chemical shift perturbation (CSP) experiments. Finally, six of the 17 new compounds have high affinity for both RpsA-CTD and its Ala438 deletion mutant. The active compounds provide new choices for targeting trans-translation in Mtb, and the analysis of the structure-activity relationships will be helpful for further structural modifications based on derivatives of 2-((hypoxanthine-2-yl)thio)acetic acid and 2-((5-hydroxylflavone-7-yl)oxy)acetamide.

Chemical space exploration based on recurrent neural networks: applications in discovering kinase inhibitors.

With the rise of artificial intelligence (AI) in drug discovery, de novo molecular generation provides new ways to explore chemical space. However, because de novo molecular generation methods rely on abundant known molecules, generated molecules may have a problem of novelty. Novelty is important in highly competitive areas of medicinal chemistry, such as the discovery of kinase inhibitors. In this study, de novo molecular generation based on recurrent neural networks was applied to discover a new chemical space of kinase inhibitors. During the application, the practicality was evaluated, and new inspiration was found. With the successful discovery of one potent Pim1 inhibitor and two lead compounds that inhibit CDK4, AI-based molecular generation shows potentials in drug discovery and development.

Development and strategies of CDK4/6 inhibitors.

Aim: CDK4/6 have critical roles in the early stage of the cell cycle. CDK2 acts later in the cell cycle and has a considerably broader range of protein substrates, some of which are essential for normal cell proliferation. Therefore, increasing the selectivity of cyclin-dependent kinase (CDK) inhibitors is critical. Methodology: In this study, we construct a versatile, specific CDK4 pharmacophore model that not only matches well with 8119 of the reported 9349 CDK4/6 inhibitors but also differentiates from the CDK2 pharmacophore. Results & Conclusion: we demonstrate the activity and selectivity determinants of CDK4/6 selective inhibitors based on the CDK4 pharmacophore model. Finally, we propose the future optimization strategy for CDK4/6 selective inhibitors, providing a theoretical basis for further research and development of CDK4/6 selective inhibitors.

Discovery of CDK4 inhibitors by convolutional neural networks.

Aim: Descriptors of molecules are important in the discovery of lead compounds. Most of these descriptors are used to represent molecular structures, although structural formulas are the most intuitive representation. Convolutional neural networks (ConvNets) are effective for managing intuitive information. Results/methodology: Convolutional neural networks (ConvNets) based on two-dimensional structural formulas were used for the preliminary screening of CDK4 inhibitors. After supervised learning of our homemade dataset, our models screened out ten approved drugs, including indocyanine green and candesartan cilexetil, with IC50 values of 2.0 and 5.2 µM, respectively. Conclusion: Depending only on intuitive information, the developed method was shown to be feasible, thus providing a new method of lead compound discovery.

An overview of neural networks for drug discovery and the inputs used.

Introduction: Artificial intelligence systems based on neural networks (NNs) find rules for drug discovery according to training molecules, but first, the molecules need to be represented in certain ways. Molecular descriptors and fingerprints have been used as inputs for artificial neural networks (ANNs) for a long time, while other ways for describing molecules are used only for storing and presenting molecules. With the development of deep learning, variants of ANNs are now able to use different kinds of inputs, which provide researchers with more choices for drug discovery. Areas covered: The authors provide a brief overview of the applications of NNs in drug discovery. Combined with the characteristics of different ways for describing molecules, corresponding methods based on NNs provide new choices for drug discovery, including de novo drug design, ligand-based drug design, and receptor-based drug design. Expert opinion: Various ways for describing molecules can be inputs of NN-based models, and these models achieve satisfactory results in metrics. Although most of the models have not been widely applied and tested in practice, they can be the basis for automatic drug discovery in the future.

Virtual screening and biological evaluation of biofilm inhibitors on dual targets in quorum sensing system.

AIM: Resistance to conventional antibiotics has spurred interest in exploring new antimicrobial strategies. Suppressing quorum sensing within biofilm is a promising antimicrobial strategy. LasR in quorum sensing system of the Gram-negative bacteria, Pseudomonas aeruginosa, directly enhances virulence and antibiotic resistance, with QscR as its indirect suppressor, so targeting both of them can synergistically take the effect. METHODOLOGY/RESULTS: An in silico protocol combining pharmacophores with molecular docking was applied. Pharmacophores of QscR agonists and LasR antagonists were prepared for preliminary screening, followed by counter-screen using a pharmacophore model of LasR agonists and molecular docking of LasR. Four compounds with novel scaffolds were confirmed as potential biofilm inhibitors with preliminary experimental data. CONCLUSION: Novel biofilm inhibitors can be found with the method.