Identification of a Selective G1-Phase Benzimidazolone Inhibitor by a Senescence-Targeted Virtual Screen Using Artificial Neural Networks

Yearning for machine learning: applications for the classification and characterisation of senescence

Senescence is a widely appreciated tumour suppressive mechanism, which acts as a barrier to cancer development by arresting cell cycle progression in response to harmful stimuli. However, senescent cell accumulation becomes deleterious in aging and contributes to a wide range of age-related pathologies. Furthermore, senescence has beneficial roles and is associated with a growing list of normal physiological processes including wound healing and embryonic development. Therefore, the biological role of senescent cells has become increasingly nuanced and complex. The emergence of sophisticated, next-generation profiling technologies, such as single-cell RNA sequencing, has accelerated our understanding of the heterogeneity of senescence, with distinct final cell states emerging within models as well as between cell types and tissues. In order to explore data sets of increasing size and complexity, the senescence field has begun to employ machine learning (ML) methodologies to probe these intricacies. Most notably, ML has been used to aid the classification of cells as senescent, as well as to characterise the final senescence phenotypes. Here, we provide a background to the principles of ML tasks, as well as some of the most commonly used methodologies from both traditional and deep ML. We focus on the application of these within the context of senescence research, by addressing the utility of ML for the analysis of data from different laboratory technologies (microscopy, transcriptomics, proteomics, methylomics), as well as the potential within senolytic drug discovery. Together, we aim to highlight both the progress and potential for the application of ML within senescence research.

A hybrid resampling algorithms SMOTE and ENN based deep learning models for identification of Marburg virus inhibitors

Background: Marburg virus (MARV) is a sporadic outbreak of a zoonotic disease that causes lethal hemorrhagic fever in humans. We propose a deep learning model with resampling techniques and predict the inhibitory activity of MARV from unknown compounds in the virtual screening process. Methodology & results: We applied resampling techniques to solve the imbalanced data problem. The classifier model comparisons revealed that the hybrid model of synthetic minority oversampling technique - edited nearest neighbor and artificial neural network (SMOTE-ENN + ANN) achieved better classification performance with 95% overall accuracy. The trained SMOTE-ENN+ANN hybrid model predicted as lead molecules; 25 out of 87,043 from ChemDiv, four out of 340 from ChEMBL anti-viral library, three out of 918 from Phytochemical database, and seven out of 419 from Natural products from NCI divsetIV, and 214 out of 1,12,267 from Natural compounds ZINC database for MARV. Conclusion: Our studies reveal that the proposed SMOTE-ENN + ANN hybrid model can improve overall accuracy more effectively and predict new lead molecules against MARV.

Synthesis of unsymmetrical 1,3-substituted-1,3-dihydro-benzimidazolones via copper-catalyzed C–N coupling under visible light

Although photoinduced copper catalysis for carbon–nitrogen (C–N) amine bond formation with alkyl/aryl halides has been developed, the potential of copper photocatalysis for the synthesis of 1,3-substituted benzimidazolones remains mostly unexplored.

Status quo and future prospects of artificial neural network from the perspective of gastroenterologists

Artificial neural networks (ANNs) are one of the primary types of artificial intelligence and have been rapidly developed and used in many fields. In recent years, there has been a sharp increase in research concerning ANNs in gastrointestinal (GI) diseases. This state-of-the-art technique exhibits excellent performance in diagnosis, prognostic prediction, and treatment. Competitions between ANNs and GI experts suggest that efficiency and accuracy might be compatible in virtue of technique advancements. However, the shortcomings of ANNs are not negligible and may induce alterations in many aspects of medical practice. In this review, we introduce basic knowledge about ANNs and summarize the current achievements of ANNs in GI diseases from the perspective of gastroenterologists. Existing limitations and future directions are also proposed to optimize ANN’s clinical potential. In consideration of barriers to interdisciplinary knowledge, sophisticated concepts are discussed using plain words and metaphors to make this review more easily understood by medical practitioners and the general public.

Applications of Virtual Screening in Bioprospecting: Facts, Shifts, and Perspectives to Explore the Chemo-Structural Diversity of Natural Products

Natural products are continually explored in the development of new bioactive compounds with industrial applications, attracting the attention of scientific research efforts due to their pharmacophore-like structures, pharmacokinetic properties, and unique chemical space. The systematic search for natural sources to obtain valuable molecules to develop products with commercial value and industrial purposes remains the most challenging task in bioprospecting. Virtual screening strategies have innovated the discovery of novel bioactive molecules assessing in silico large compound libraries, favoring the analysis of their chemical space, pharmacodynamics, and their pharmacokinetic properties, thus leading to the reduction of financial efforts, infrastructure, and time involved in the process of discovering new chemical entities. Herein, we discuss the computational approaches and methods developed to explore the chemo-structural diversity of natural products, focusing on the main paradigms involved in the discovery and screening of bioactive compounds from natural sources, placing particular emphasis on artificial intelligence, cheminformatics methods, and big data analyses.

Discovery of novel dual adenosine A1/A2A receptor antagonists using deep learning, pharmacophore modeling and molecular docking

Adenosine receptors (ARs) have been demonstrated to be potential therapeutic targets against Parkinson's disease (PD). In the present study, we describe a multistage virtual screening approach that identifies dual adenosine A1 and A2A receptor antagonists using deep learning, pharmacophore models, and molecular docking methods. Nineteen hits from the ChemDiv library containing 1,178,506 compounds were selected and further tested by in vitro assays (cAMP functional assay and radioligand binding assay); of these hits, two compounds (C8 and C9) with 1,2,4-triazole scaffolds possessing the most potent binding affinity and antagonistic activity for A1/A2A ARs at the nanomolar level (pKi of 7.16-7.49 and pIC50 of 6.31-6.78) were identified. Further molecular dynamics (MD) simulations suggested similarly strong binding interactions of the complexes between the A1/A2A ARs and two compounds (C8 and C9). Notably, the 1,2,4-triazole derivatives (compounds C8 and C9) were identified as the most potent dual A1/A2A AR antagonists in our study and could serve as a basis for further development. The effective multistage screening approach developed in this study can be utilized to identify potent ligands for other drug targets.

A Novel Pyrazolopyrimidine Ligand of Human PGK1 and Stress Sensor DJ1 Modulates the Shelterin Complex and Telomere Length Regulation

Telomere signaling and metabolic dysfunction are hallmarks of cell aging. New agents targeting these processes might provide therapeutic opportunities, including chemoprevention strategies against cancer predisposition. We report identification and characterization of a pyrazolopyrimidine compound series identified from screens focused on cell immortality and whose targets are glycolytic kinase PGK1 and oxidative stress sensor DJ1. We performed structure–activity studies on the series to develop a photoaffinity probe to deconvolute the cellular targets. In vitro binding and structural analyses confirmed these targets, suggesting that PGK1/DJ1 interact, which we confirmed by immunoprecipitation. Glucose homeostasis and oxidative stress are linked to telomere signaling and exemplar compound CRT0063465 blocked hypoglycemic telomere shortening. Intriguingly, PGK1 and DJ1 bind to TRF2 and telomeric DNA. Compound treatment modulates these interactions and also affects Shelterin complex composition, while conferring cellular protection from cytotoxicity due to bleomycin and desferroxamine. These results demonstrate therapeutic potential of the compound series.

Deep learning and virtual drug screening

Current drug development is still costly and slow given tremendous technological advancements in drug discovery and medicinal chemistry. Using machine learning (ML) to virtually screen compound libraries promises to fix this for generating drug leads more efficiently and accurately. Herein, we explain the broad basics and integration of both virtual screening (VS) and ML. We then discuss artificial neural networks (ANNs) and their usage for VS. The ANN is emerging as the dominant classifier for ML in general, and has proven its utility for both structure-based and ligand-based VS. Techniques such as dropout, multitask learning and convolution improve the performance of ANNs and enable them to take on chemical meaning when learning about the drug-target-binding activity of compounds.

A 'synthetic-sickness' screen for senescence re-engagement targets in mutant cancer backgrounds

Senescence is a universal barrier to immortalisation and tumorigenesis. As such, interest in the use of senescence-induction in a therapeutic context has been gaining momentum in the past few years; however, senescence and immortalisation remain underserved areas for drug discovery owing to a lack of robust senescence inducing agents and an incomplete understanding of the signalling events underlying this complex process. In order to address this issue we undertook a large-scale morphological siRNA screen for inducers of senescence phenotypes in the human melanoma cell line A375P. Following rescreen and validation in a second cancer cell line, HCT116 colorectal carcinoma, a panel of 16 of the most robust hits were selected for further validation based on significance and the potential to be targeted by drug-like molecules. Using secondary assays for detection of senescence biomarkers p21, 53BP1 and senescence associated beta-galactosidase (SAβGal) in a panel of HCT116 cell lines carrying cancer-relevant mutations, we show that partial senescence phenotypes can be induced to varying degrees in a context dependent manner, even in the absence of p21 or p53 expression. However, proliferation arrest varied among genetic backgrounds with predominantly toxic effects in p21 null cells, while cells lacking PI3K mutation failed to arrest. Furthermore, we show that the oncogene ECT2 induces partial senescence phenotypes in all mutant backgrounds tested, demonstrating a dependence on activating KRASG13D for growth suppression and a complete senescence response. These results suggest a potential mechanism to target mutant KRAS signalling through ECT2 in cancers that are reliant on activating KRAS mutations and remain refractory to current treatments.

The Virtual Screening of the Drug Protein with a Few Crystal Structures Based on the Adaboost-SVM

Using the theory of machine learning to assist the virtual screening (VS) has been an effective plan. However, the quality of the training set may reduce because of mixing with the wrong docking poses and it will affect the screening efficiencies. To solve this problem, we present a method using the ensemble learning to improve the support vector machine to process the generated protein-ligand interaction fingerprint (IFP). By combining multiple classifiers, ensemble learning is able to avoid the limitations of the single classifier's performance and obtain better generalization. According to the research of virtual screening experiment with SRC and Cathepsin K as the target, the results show that the ensemble learning method can effectively reduce the error because the sample quality is not high and improve the effect of the whole virtual screening process.