Search

Given their high affinity and specificity for a range of macromolecules, antibodies are widely used in the treatment of autoimmune diseases, cancers, inflammatory diseases, and Alzheimer's disease (AD). Traditional experimental methods are time-consuming, expensive, and labor-intensive. Recent advances in artificial intelligence (AI) technologies provide complementary methods that can reduce the time and costs required for antibody design by minimizing failures and increasing the success rate of experimental tests. In this review, we scrutinize the plethora of AI-driven methodologies that have been deployed over the past 4 years for modeling antibody structures, predicting antibody-antigen interactions, optimizing antibody affinity, and generating novel antibody candidates. We also briefly address the challenges faced in integrating AI-based models with traditional antibody discovery pipelines and highlight the potential future directions in this burgeoning field., (Copyright © 2024 Elsevier Ltd. All rights reserved.)

Although drugs targeting the orthosteric binding site of cannabinoid receptors (CBRs) have several therapeutic effects on human physiological and pathological conditions, they can also cause serious adverse effects. Only a few orthosteric ligands have successfully passed clinical trials. Recently, allosteric modulation has become a novel option for drug discovery, with fewer adverse effects and the potential to avoid drug overdose. In this review, we highlight novel findings related to the drug discovery of allosteric modulators (AMs) targeting CBRs. We summarize newly synthesized AMs and the reported/predicted allosteric binding sites. We also discuss the structural determinants of the AMs binding as well as the molecular mechanism of CBR allostery., (Copyright © 2023 Elsevier Ltd. All rights reserved.)

Pancreatic ductal adenocarcinoma (PDAC) is characterized by heightened autophagy and systemic immune dysfunction. Modest improvements in clinical outcomes have been demonstrated in completed clinical trials targeting autophagy with combination hydroxychloroquine (HCQ) and chemotherapy. Recent mechanistic insights into the role of autophagy-dependent immune evasion have prompted the need for more precise and druggable targets of autophagy inhibition. Sequestosome-1 (SQSTM-1) is a multidomain scaffold protein with well-established roles in autophagy, tumor necrosis factor alpha (TNFα)- and NF-κB-related signaling pathways. SQSTM1 overexpression is frequently observed in PDAC, correlating with clinical stage and outcome. Given the unique molecular structure of SQSTM-1 and its diverse activity, identifying means of limiting SQSTM-1-dependent autophagy to promote an effective immune response in PDAC could be a promising treatment strategy., (Copyright © 2021 Elsevier Ltd. All rights reserved.)

The de novo design of molecular structures using deep learning generative models introduces an encouraging solution to drug discovery in the face of the continuously increased cost of new drug development. From the generation of original texts, images, and videos, to the scratching of novel molecular structures the creativity of deep learning generative models exhibits the height machine intelligence can achieve. The purpose of this paper is to review the latest advances in generative chemistry which relies on generative modeling to expedite the drug discovery process. This review starts with a brief history of artificial intelligence in drug discovery to outline this emerging paradigm. Commonly used chemical databases, molecular representations, and tools in cheminformatics and machine learning are covered as the infrastructure for generative chemistry. The detailed discussions on utilizing cutting-edge generative architectures, including recurrent neural network, variational autoencoder, adversarial autoencoder, and generative adversarial network for compound generation are focused. Challenges and future perspectives follow.

Designing covalent allosteric modulators brings new opportunities to the field of drug discovery towards G-protein-coupled receptors (GPCRs). Targeting an allosteric binding pocket can allow a modulator to have protein subtype selectivity and low drug resistance. Utilizing covalent warheads further enables the modulator to increase the binding potency and extend the duration of action. This review starts with GPCR allosteric modulation to discuss the structural biology of allosteric binding pockets, the different types of allosteric modulators, as well as the advantages of employing allosteric modulation. This is followed by a discussion on covalent modulators to clarify how covalent ligands can benefit the receptor modulation and to illustrate moieties that can commonly be used as covalent warheads. Finally, case studies are presented on designing class A, B, and C GPCR covalent allosteric modulators to demonstrate successful stories on combining allosteric modulation and covalent binding. Limitations and future perspectives are also covered., Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper., (Copyright © 2020 Elsevier Masson SAS. All rights reserved.)

A deep convolutional generative adversarial network (dcGAN) model was developed in this study to screen and design target-specific novel compounds for cannabinoid receptors. In the adversarial process of training, two models, the discriminator D and the generator G, are iteratively trained. D is trained to discover the hidden patterns among the input data to have the accurate discrimination of the authentic compounds and the "fake" compounds generated by G; G is trained to generate "fake" compounds to fool the well-trained D by optimizing the weights for matrix multiplication of data sampling. In order to determine the appropriate architecture and the input data structure for the involved convolutional neural networks (CNNs), the combinations of various network architectures and molecular fingerprints were explored. Well-developed CNN models including LeNet-5, AlexNet, ZFNet, and VGGNet were investigated. Four types of fingerprints, including MACCS, ECFP6, AtomPair, and AtomPair Count, were calculated to describe the small molecules with diverse structural characteristics. The limitation of generating fingerprints as output remains that the concrete molecular structures cannot be converted directly, while the generative models with convolutional networks provide promising opportunities to the screening of molecules and rational modifications afterward. This study demonstrated how computer-aided drug discovery could benefit from the recent advances in deep learning.

Traumatic brain injury (TBI) is associated with high mortality and morbidity. Though the death rate of initial trauma has dramatically decreased, no drug has been developed to effectively limit the progression of the secondary injury caused by TBI. TBI appears to be a predisposing risk factor for Alzheimer's disease (AD), whereas the molecular mechanisms remain unknown. In this study, we have conducted a research investigation of computational chemogenomics systems pharmacology (CSP) to identify potential drug targets for TBI treatment. TBI-induced transcriptional profiles were compared with those induced by genetic or chemical perturbations, including drugs in clinical trials for TBI treatment. The protein-protein interaction network of these predicted targets were then generated for further analyses. Some protein targets when perturbed, exhibit inverse transcriptional profiles in comparison with the profiles induced by TBI, and they were recognized as potential therapeutic targets for TBI. Drugs acting on these targets are predicted to have the potential for TBI treatment if they can reverse the TBI-induced transcriptional profiles that lead to secondary injury. In particular, our results indicated that TRPV4, NEUROD1, and HPRT1 were among the top therapeutic target candidates for TBI, which are congruent with literature reports. Our analyses also suggested the strong associations between TBI and AD, as perturbations on AD-related genes, such as APOE, APP, PSEN1, and MAPT, can induce similar gene expression patterns as those of TBI. To the best of our knowledge, this is the first CSP-based gene expression profile analyses for predicting TBI-related drug targets, and the findings could be used to guide the design of new drugs targeting the secondary injury caused by TBI.

Over the last decade, deep learning (DL) methods have been extremely successful and widely used to develop artificial intelligence (AI) in almost every domain, especially after it achieved its proud record on computational Go. Compared to traditional machine learning (ML) algorithms, DL methods still have a long way to go to achieve recognition in small molecular drug discovery and development. And there is still lots of work to do for the popularization and application of DL for research purpose, e.g., for small molecule drug research and development. In this review, we mainly discussed several most powerful and mainstream architectures, including the convolutional neural network (CNN), recurrent neural network (RNN), and deep auto-encoder networks (DAENs), for supervised learning and nonsupervised learning; summarized most of the representative applications in small molecule drug design; and briefly introduced how DL methods were used in those applications. The discussion for the pros and cons of DL methods as well as the main challenges we need to tackle were also emphasized.

Molecular docking is widely applied to computer-aided drug design and has become relatively mature in the recent decades. Application of docking in modeling varies from single lead compound optimization to large-scale virtual screening. The performance of molecular docking is highly dependent on the protein structures selected. It is especially challenging for large-scale target prediction research when multiple structures are available for a single target. Therefore, we have established ProSelection, a docking preferred-protein selection algorithm, in order to generate the proper structure subset(s). By the ProSelection algorithm, protein structures of "weak selectors" are filtered out whereas structures of "strong selectors" are kept. Specifically, the structure which has a good statistical performance of distinguishing active ligands from inactive ligands is defined as a strong selector. In this study, 249 protein structures of 14 autophagy-related targets are investigated. Surflex-dock was used as the docking engine to distinguish active and inactive compounds against these protein structures. Both t test and Mann-Whitney U test were used to distinguish the strong from the weak selectors based on the normality of the docking score distribution. The suggested docking score threshold for active ligands (SDA) was generated for each strong selector structure according to the receiver operating characteristic (ROC) curve. The performance of ProSelection was further validated by predicting the potential off-targets of 43 U.S. Federal Drug Administration approved small molecule antineoplastic drugs. Overall, ProSelection will accelerate the computational work in protein structure selection and could be a useful tool for molecular docking, target prediction, and protein-chemical database establishment research.

Given the capacity of self-renewal and multilineage differentiation, stem cells are promising sources for use in regenerative medicines as well as in the clinical treatment of certain hematological malignancies and degenerative diseases. Complex networks of cellular signaling pathways largely determine stem cell fate and function. Small molecules that modulate these pathways can provide important biological and pharmacological insights. However, it is still challenging to identify the specific protein targets of these compounds, to explore the changes in stem cell phenotypes induced by compound treatment and to ascertain compound mechanisms of action. To facilitate stem cell related small molecule study and provide a better understanding of the associated signaling pathways, we have constructed a comprehensive domain-specific chemogenomics resource, called StemCellCKB ( http://www.cbligand.org/StemCellCKB/ ). This new cloud-computing platform describes the chemical molecules, genes, proteins, and signaling pathways implicated in stem cell regulation. StemCellCKB is also implemented with web applications designed specifically to aid in the identification of stem cell relevant protein targets, including TargetHunter, a machine-learning algorithm for predicting small molecule targets based on molecular fingerprints, and HTDocking, a high-throughput docking module for target prediction and systems-pharmacology analyses. We have systematically tested StemCellCKB to verify data integrity. Target-prediction accuracy has also been validated against the reported known target/compound associations. This proof-of-concept example demonstrates that StemCellCKB can (1) accurately predict the macromolecular targets of existing stem cell modulators and (2) identify novel small molecules capable of probing stem cell signaling mechanisms, for use in systems-pharmacology studies. StemCellCKB facilitates the exploration and exchange of stem cell chemogenomics data among members of the broader research community.

Cannabinoid receptor 2 (CB2), a G protein-coupled receptor (GPCR), is a promising target for the treatment of neuropathic pain, osteoporosis, immune system, cancer, and drug abuse. The lack of an experimental three-dimensional CB2 structure has hindered not only the development of studies of conformational differences between the inactive and active CB2 but also the rational discovery of novel functional compounds targeting CB2. In this work, we constructed models of both inactive and active CB2 by homology modeling. Then we conducted two comparative 100 ns molecular dynamics (MD) simulations on the two systems-the active CB2 bound with both the agonist and G protein and the inactive CB2 bound with inverse agonist-to analyze the conformational difference of CB2 proteins and the key residues involved in molecular recognition. Our results showed that the inactive CB2 and the inverse agonist remained stable during the MD simulation. However, during the MD simulations, we observed dynamical details about the breakdown of the "ionic lock" between R131(3.50) and D240(6.30) as well as the outward/inward movements of transmembrane domains of the active CB2 that bind with G proteins and agonist (TM5, TM6, and TM7). All of these results are congruent with the experimental data and recent reports. Moreover, our results indicate that W258(6.48) in TM6 and residues in TM4 (V164(4.56)-L169(4.61)) contribute greatly to the binding of the agonist on the basis of the binding energy decomposition, while residues S180-F183 in extracellular loop 2 (ECL2) may be of importance in recognition of the inverse agonist. Furthermore, pharmacophore modeling and virtual screening were carried out for the inactive and active CB2 models in parallel. Among all 10 hits, two compounds exhibited novel scaffolds and can be used as novel chemical probes for future studies of CB2. Importantly, our studies show that the hits obtained from the inactive CB2 model mainly act as inverse agonist(s) or neutral antagonist(s) at low concentration. Moreover, the hit from the active CB2 model also behaves as a neutral antagonist at low concentration. Our studies provide new insight leading to a better understanding of the structural and conformational differences between two states of CB2 and illuminate the effects of structure on virtual screening and drug design.

Target identification of the known bioactive compounds and novel synthetic analogs is a very important research field in medicinal chemistry, biochemistry, and pharmacology. It is also a challenging and costly step towards chemical biology and phenotypic screening. In silico identification of potential biological targets for chemical compounds offers an alternative avenue for the exploration of ligand-target interactions and biochemical mechanisms, as well as for investigation of drug repurposing. Computational target fishing mines biologically annotated chemical databases and then maps compound structures into chemogenomical space in order to predict the biological targets. We summarize the recent advances and applications in computational target fishing, such as chemical similarity searching, data mining/machine learning, panel docking, and the bioactivity spectral analysis for target identification. We then described in detail a new web-based target prediction tool, TargetHunter (http://www.cbligand.org/TargetHunter). This web portal implements a novel in silico target prediction algorithm, the Targets Associated with its MOst SImilar Counterparts, by exploring the largest chemogenomical databases, ChEMBL. Prediction accuracy reached 91.1% from the top 3 guesses on a subset of high-potency compounds from the ChEMBL database, which outperformed a published algorithm, multiple-category models. TargetHunter also features an embedded geography tool, BioassayGeoMap, developed to allow the user easily to search for potential collaborators that can experimentally validate the predicted biological target(s) or off target(s). TargetHunter therefore provides a promising alternative to bridge the knowledge gap between biology and chemistry, and significantly boost the productivity of chemogenomics researchers for in silico drug design and discovery.

In this manuscript, we have reported a novel 2D fingerprint-based artificial neural network QSAR (FANN-QSAR) method in order to effectively predict biological activities of structurally diverse chemical ligands. Three different types of fingerprints, namely, ECFP6, FP2 and MACCS, were used in FANN-QSAR algorithm development, and FANN-QSAR models were compared to known 3D and 2D QSAR methods using five data sets previously reported. In addition, the derived models were used to predict GPCR cannabinoid ligand binding affinities using our manually curated cannabinoid ligand database containing 1699 structurally diverse compounds with reported cannabinoid receptor subtype CB(2) activities. To demonstrate its useful applications, the established FANN-QSAR algorithm was used as a virtual screening tool to search a large NCI compound database for lead cannabinoid compounds, and we have discovered several compounds with good CB(2) binding affinities ranging from 6.70 nM to 3.75 μM. To the best of our knowledge, this is the first report for a fingerprint-based neural network approach validated with a successful virtual screening application in identifying lead compounds. The studies proved that the FANN-QSAR method is a useful approach to predict bioactivities or properties of ligands and to find novel lead compounds for drug discovery research.

Delineating the fingerprint or feature vector of a receptor/protein will facilitate the structural and biological studies, as well as the rational design and development of drugs with high affinities and selectivity. However, protein is complicated by its different functional regions that can bind to some of its protein partner(s), substrate(s), orthosteric ligand(s) or allosteric modulator(s) where cogent methods like molecular fingerprints do not work well. We here elaborate a scoring-function-based computing protocol Molecular Complex Characterizing System to help characterize the binding feature of protein–ligand complexes. Based on the reported receptor-ligand interactions, we first quantitate the energy contribution of each individual residue which may be an alternative of MD-based energy decomposition. We then construct a vector for the energy contribution to represent the pattern of the ligand recognition at a receptor and qualitatively analyze the matching level with other receptors. Finally, the energy contribution vector is explored for extensive use in similarity and clustering. The present work provides a new approach to cluster proteins, a perspective counterpart for determining the protein characteristics in the binding, and an advanced screening technique where molecular docking is applicable. [ABSTRACT FROM AUTHOR]

Design and generation of high-quality target- and scaffold-specific small molecules is an important strategy for the discovery of unique and potent bioactive drug molecules. To achieve this goal, authors have developed the deep-learning molecule generation model (DeepMGM) and applied it for the de novo molecular generation of scaffold-focused small-molecule libraries. In this study, a recurrent neural network (RNN) using long short-term memory (LSTM) units was trained with drug-like molecules to result in a general model (g-DeepMGM). Sampling practices on indole and purine scaffolds illustrate the feasibility of creating scaffold-focused chemical libraries based on machine intelligence. Subsequently, a target-specific model (t-DeepMGM) for cannabinoid receptor 2 (CB2) was constructed following the transfer learning process of known CB2 ligands. Sampling outcomes can present similar properties to the reported active molecules. Finally, a discriminator was trained and attached to the DeepMGM to result in an in silico molecular design-test circle. Medicinal chemistry synthesis and biological validation was performed to further investigate the generation outcome, showing that XIE9137 was identified as a potential allosteric modulator of CB2. This study demonstrates how recent progress in deep learning intelligence can benefit drug discovery, especially in de novo molecular design and chemical library generation. [ABSTRACT FROM AUTHOR]