De novo design and bioactivity prediction of SARS-CoV-2 main protease inhibitors using recurrent neural network-based transfer learning

Identification of SARS-CoV-2 Mpro inhibitors through deep reinforcement learning for de novo drug design and computational chemistry approaches

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused a global pandemic of coronavirus disease (COVID-19) since its emergence in December 2019. As of January 2024, there has been over 774 million reported cases and 7 million deaths worldwide. While vaccination efforts have been successful in reducing the severity of the disease and decreasing the transmission rate, the development of effective therapeutics against SARS-CoV-2 remains a critical need. The main protease (Mpro) of SARS-CoV-2 is an essential enzyme required for viral replication and has been identified as a promising target for drug development. In this study, we report the identification of novel Mpro inhibitors, using a combination of deep reinforcement learning for de novo drug design with 3D pharmacophore/shape-based alignment and privileged fragment match count scoring components followed by hit expansions and molecular docking approaches. Our experimentally validated results show that 3 novel series exhibit potent inhibitory activity against SARS-CoV-2 Mpro, with IC50 values ranging from 1.3 μM to 2.3 μM and a high degree of selectivity. These findings represent promising starting points for the development of new antiviral therapies against COVID-19.

SchNetPack Hyperparameter Optimization for a More Reliable Top Docking Scores Prediction

Options to improve the extrapolation power of the neural network designed using the SchNetPack package with respect to top docking scores prediction are presented. It is shown that hyperparameter tuning of the atomistic model representation (in the schnetpack.representation) improves the prediction of the top scoring compounds, which have characteristically a low incidence in randomized data sets for training of machine learning models. The prediction robustness is evaluated according to the mean square error (MSE) and the entropy of the average loss landscape decrease. Admittedly, the improvement of the top scoring compounds' prediction accuracy comes with the penalty of worsening the overall prediction power. It is revealed that the most impactful hyperparameter is the cutoff (5 Å is reported as the optimal choice). Other parameters (e.g., number of radial basis functions, number of interaction layers of the neural network, feature vector size or its batch size) are found to not affect the prediction robustness of the top scoring compounds in any comparable way relative to the cutoff. The MSE of the best docking score prediction (below -13 kcal/mol) improves from ca. 3.5 to 0.9 kcal/mol, while the prediction of less potent compounds (-13 to -11 kcal/mol) shows a lesser improvement, i.e., a decrease of MSE from 1.6 to 1.3 kcal/mol. Additionally, oversampling and undersampling of the training set with respect to the top scoring compounds' abundance is presented. The results indicate that the cutoff choice performs better than over- or undersampling of the training set, with undersampling performing better than oversampling.

De novo generation of dual-target ligands for the treatment of SARS-CoV-2 using deep learning, virtual screening, and molecular dynamic simulations

De novo generation of molecules with the necessary features offers a promising opportunity for artificial intelligence, such as deep generative approaches. However, creating novel compounds having biological activities toward two distinct targets continues to be a very challenging task. In this study, we develop a unique computational framework for the de novo synthesis of bioactive compounds directed at two predetermined therapeutic targets. This framework is referred to as the dual-target ligand generative network. Our approach uses a stochastic policy to explore chemical spaces called a sequence-based simple molecular input line entry system (SMILES) generator. The steps in the high-level workflow would be to gather and prepare the training data for both targets' molecules, build a neural network model and train it to make molecules, create new molecules using generative AI, and then virtually screen the newly validated molecules against the SARS-CoV-2 PLpro and 3CLpro drug targets. Results shows that novel molecules generated have higher binding affinity with both targets than the conventional drug i.e. Remdesivir being used for the treatment of SARS-CoV-2.Communicated by Ramaswamy H. Sarma.

Machine learning combines atomistic simulations to predict SARS-CoV-2 Mpro inhibitors from natural compounds

To date, the COVID-19 pandemic has still been infectious around the world, continuously causing social and economic damage on a global scale. One of the most important therapeutic targets for the treatment of COVID-19 is the main protease (Mpro) of SARS-CoV-2. In this study, we combined machine-learning (ML) model with atomistic simulations to computationally search for highly promising SARS-CoV-2 Mpro inhibitors from the representative natural compounds of the National Cancer Institute (NCI) Database. First, the trained ML model was used to scan the library quickly and reliably for possible Mpro inhibitors. The ML output was then confirmed using atomistic simulations integrating molecular docking and molecular dynamic simulations with the linear interaction energy scheme. The results turned out to show that there was evidently good agreement between ML and atomistic simulations. Ten substances were proposed to be able to inhibit SARS-CoV-2 Mpro. Seven of them have high-nanomolar affinity and are very potential inhibitors. The strategy has been proven to be reliable and appropriate for fast prediction of SARS-CoV-2 Mpro inhibitors, benefiting for new emerging SARS-CoV-2 variants in the future accordingly.

Cluster energy prediction based on multiple strategy fusion whale optimization algorithm and light gradient boosting machine

BACKGROUND

Clusters, a novel hierarchical material structure that emerges from atoms or molecules, possess unique reactivity and catalytic properties, crucial in catalysis, biomedicine, and optoelectronics. Predicting cluster energy provides insights into electronic structure, magnetism, and stability. However, the structure of clusters and their potential energy surface is exceptionally intricate. Searching for the global optimal structure (the lowest energy) among these isomers poses a significant challenge. Currently, modelling cluster energy predictions with traditional machine learning methods has several issues, including reliance on manual expertise, slow computation, heavy computational resource demands, and less efficient parameter tuning.

RESULTS

This paper introduces a predictive model for the energy of a gold cluster comprising twenty atoms (referred to as Au20 cluster). The model integrates the Multiple Strategy Fusion Whale Optimization Algorithm (MSFWOA) with the Light Gradient Boosting Machine (LightGBM), resulting in the MSFWOA-LightGBM model. This model employs the Coulomb matrix representation and eigenvalue solution methods for feature extraction. Additionally, it incorporates the Tent chaotic mapping, cosine convergence factor, and inertia weight updating strategy to optimize the Whale Optimization Algorithm (WOA), leading to the development of MSFWOA. Subsequently, MSFWOA is employed to optimize the parameters of LightGBM for supporting the energy prediction of Au20 cluster.

CONCLUSIONS

The experimental results show that the most stable Au20 cluster structure is a regular tetrahedron with the lowest energy, displaying tight and uniform atom distribution, high geometric symmetry. Compared to other models, the MSFWOA-LightGBM model excels in accuracy and correlation, with MSE, RMSE, and R2 values of 0.897, 0.947, and 0.879, respectively. Additionally, the MSFWOA-LightGBM model possesses outstanding scalability, offering valuable insights for material design, energy storage, sensing technology, and biomedical imaging, with the potential to drive research and development in these areas.

VGAE-MCTS: A New Molecular Generative Model Combining the Variational Graph Auto-Encoder and Monte Carlo Tree Search

Molecular generation is crucial for advancing drug discovery, materials science, and chemical exploration. It expedites the search for new drug candidates, facilitates tailored material creation, and enhances our understanding of molecular diversity. By employing artificial intelligence techniques such as molecular generative models based on molecular graphs, researchers have tackled the challenge of identifying efficient molecules with desired properties. Here, we propose a new molecular generative model combining a graph-based deep neural network and a reinforcement learning technique. We evaluated the validity, novelty, and optimized physicochemical properties of the generated molecules. Importantly, the model explored uncharted regions of chemical space, allowing for the efficient discovery and design of new molecules. This innovative approach has considerable potential to revolutionize drug discovery, materials science, and chemical research for accelerating scientific innovation. By leveraging advanced techniques and exploring previously unexplored chemical spaces, this study offers promising prospects for the efficient discovery and design of new molecules in the field of drug development.

Generation of focused drug molecule library using recurrent neural network

CONTEXT

With the wide application of deep learning in drug research and development, de novo molecular design methods based on recurrent neural network (RNN) have strong advantages in drug molecule generation. The RNN model can be used to learn the internal chemical structure of molecules, which is similar to a natural language processing task. Although techniques for generating target-specific molecular libraries based on RNN models are mature, research related to drug design and screening continues around the clock. Research based on de novo drug design methods to generate larger quantities of valid compounds is necessary.

METHODS

In this study, a molecular generation model based on RNN was designed, which abandoned the traditional way of stacked RNN and introduced the Nested long short-term memory network structure. To enrich the library of focused molecules for specific targets, we fine-tuned the model using active molecules from novel coronavirus pneumonia and screened the molecules using machine learning models. Following rigorous screening, the selected molecules underwent molecular docking with the SARS-CoV-2 M-pro receptor using AutoDock2.4 to identify the top 3 potential inhibitors. Subsequently, 100-ns molecular dynamics simulations were conducted using Amber22. Molecule parameterization involved the GAFF2 force field, while the proteins were modeled using the ff19SB force field, with solvation facilitated by a truncated octahedral TIP3P solvent environment. Upon completion of molecular dynamics simulations, stability of ligand-protein complexes was assessed by analysis of RMSD, H-bonds, and MM-GBSA. Reasonable results prove that the model can complete the task of de novo drug design and has the potential to be ideal drug molecules.

Bio-activity prediction of drug candidate compounds targeting SARS-Cov-2 using machine learning approaches

The SARS-CoV-2 3CLpro protein is one of the key therapeutic targets of interest for COVID-19 due to its critical role in viral replication, various high-quality protein crystal structures, and as a basis for computationally screening for compounds with improved inhibitory activity, bioavailability, and ADMETox properties. The ChEMBL and PubChem database contains experimental data from screening small molecules against SARS-CoV-2 3CLpro, which expands the opportunity to learn the pattern and design a computational model that can predict the potency of any drug compound against coronavirus before in-vitro and in-vivo testing. In this study, Utilizing several descriptors, we evaluated 27 machine learning classifiers. We also developed a neural network model that can correctly identify bioactive and inactive chemicals with 91% accuracy, on CheMBL data and 93% accuracy on combined data on both CheMBL and Pubchem. The F1-score for inactive and active compounds was 93% and 94%, respectively. SHAP (SHapley Additive exPlanations) on XGB classifier to find important fingerprints from the PaDEL descriptors for this task. The results indicated that the PaDEL descriptors were effective in predicting bioactivity, the proposed neural network design was efficient, and the Explanatory factor through SHAP correctly identified the important fingertips. In addition, we validated the effectiveness of our proposed model using a large dataset encompassing over 100,000 molecules. This research employed various molecular descriptors to discover the optimal one for this task. To evaluate the effectiveness of these possible medications against SARS-CoV-2, more in-vitro and in-vivo research is required.

Staying Ahead of the Game: How SARS-CoV-2 has Accelerated the Application of Machine Learning in Pandemic Management

In recent years, machine learning (ML) techniques have garnered considerable interest for their potential use in accelerating the rate of drug discovery. With the emergence of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic, the utilization of ML has become even more crucial in the search for effective antiviral medications. The pandemic has presented the scientific community with a unique challenge, and the rapid identification of potential treatments has become an urgent priority. Researchers have been able to accelerate the process of identifying drug candidates, repurposing existing drugs, and designing new compounds with desirable properties using machine learning in drug discovery. To train predictive models, ML techniques in drug discovery rely on the analysis of large datasets, including both experimental and clinical data. These models can be used to predict the biological activities, potential side effects, and interactions with specific target proteins of drug candidates. This strategy has proven to be an effective method for identifying potential coronavirus disease 2019 (COVID-19) and other disease treatments. This paper offers a thorough analysis of the various ML techniques implemented to combat COVID-19, including supervised and unsupervised learning, deep learning, and natural language processing. The paper discusses the impact of these techniques on pandemic drug development, including the identification of potential treatments, the understanding of the disease mechanism, and the creation of effective and safe therapeutics. The lessons learned can be applied to future outbreaks and drug discovery initiatives.

Practical guidelines for the use of gradient boosting for molecular property prediction

Decision tree ensembles are among the most robust, high-performing and computationally efficient machine learning approaches for quantitative structure-activity relationship (QSAR) modeling. Among them, gradient boosting has recently garnered particular attention, for its performance in data science competitions, virtual screening campaigns, and bioactivity prediction. However, different variants of gradient boosting exist, the most popular being XGBoost, LightGBM and CatBoost. Our study provides the first comprehensive comparison of these approaches for QSAR. To this end, we trained 157,590 gradient boosting models, which were evaluated on 16 datasets and 94 endpoints, comprising 1.4 million compounds in total. Our results show that XGBoost generally achieves the best predictive performance, while LightGBM requires the least training time, especially for larger datasets. In terms of feature importance, the models surprisingly rank molecular features differently, reflecting differences in regularization techniques and decision tree structures. Thus, expert knowledge must always be employed when evaluating data-driven explanations of bioactivity. Furthermore, our results show that the relevance of each hyperparameter varies greatly across datasets and that it is crucial to optimize as many hyperparameters as possible to maximize the predictive performance. In conclusion, our study provides the first set of guidelines for cheminformatics practitioners to effectively train, optimize and evaluate gradient boosting models for virtual screening and QSAR applications.

In Silico Identification of Potential Inhibitors of the SARS-CoV-2 Main Protease among a PubChem Database of Avian Infectious Bronchitis Virus 3CLPro Inhibitors

Remarkable structural homologies between the main proteases of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the avian infectious bronchitis virus (IBV) were revealed by comparative amino-acid sequence and 3D structural alignment. Assessing whether reported IBV 3CLPro inhibitors could also interact with SARS-CoV-2 has been undertaken in silico using a PubChem BioAssay database of 388 compounds active on the avian infectious bronchitis virus 3C-like protease. Docking studies of this database on the SARS-CoV-2 protease resulted in the identification of four covalent inhibitors targeting the catalytic cysteine residue and five non-covalent inhibitors for which the binding was further investigated by molecular dynamics (MD) simulations. Predictive ADMET calculations on the nine compounds suggest promising pharmacokinetic properties.

AI-Driven Discovery of SARS-CoV-2 Main Protease Fragment-like Inhibitors with Antiviral Activity In Vitro

SARS-CoV-2 is the causative agent of COVID-19 and is responsible for the current global pandemic. The viral genome contains 5 major open reading frames of which the largest ORF1ab codes for two polyproteins, pp1ab and pp1a, which are subsequently cleaved into 16 nonstructural proteins (nsp) by two viral cysteine proteases encoded within the polyproteins. The main protease (Mpro, nsp5) cleaves the majority of the nsp's, making it essential for viral replication and has been successfully targeted for the development of antivirals. The first oral Mpro inhibitor, nirmatrelvir, was approved for treatment of COVID-19 in late December 2021 in combination with ritonavir as Paxlovid. Increasing the arsenal of antivirals and development of protease inhibitors and other antivirals with a varied mode of action remains a priority to reduce the likelihood for resistance emerging. Here, we report results from an artificial intelligence-driven approach followed by in vitro validation, allowing the identification of five fragment-like Mpro inhibitors with IC50 values ranging from 1.5 to 241 μM. The three most potent molecules (compounds 818, 737, and 183) were tested against SARS-CoV-2 by in vitro replication in Vero E6 and Calu-3 cells. Compound 818 was active in both cell models with an EC50 value comparable to its measured IC50 value. On the other hand, compounds 737 and 183 were only active in Calu-3, a preclinical model of respiratory cells, showing selective indexes twice as high as those for compound 818. We also show that our in silico methodology was successful in identifying both reversible and covalent inhibitors. For instance, compound 818 is a reversible chloromethylamide analogue of 8-methyl-γ-carboline, while compound 737 is an N-pyridyl-isatin that covalently inhibits Mpro. Given the small molecular weights of these fragments, their high binding efficiency in vitro and efficacy in blocking viral replication, these compounds represent good starting points for the development of potent lead molecules targeting the Mpro of SARS-CoV-2.

Magicmol: a light-weighted pipeline for drug-like molecule evolution and quick chemical space exploration

The flourishment of machine learning and deep learning methods has boosted the development of cheminformatics, especially regarding the application of drug discovery and new material exploration. Lower time and space expenses make it possible for scientists to search the enormous chemical space. Recently, some work combined reinforcement learning strategies with recurrent neural network (RNN)-based models to optimize the property of generated small molecules, which notably improved a batch of critical factors for these candidates. However, a common problem among these RNN-based methods is that several generated molecules have difficulty in synthesizing despite owning higher desired properties such as binding affinity. However, RNN-based framework better reproduces the molecule distribution among the training set than other categories of models during molecule exploration tasks. Thus, to optimize the whole exploration process and make it contribute to the optimization of specified molecules, we devised a light-weighted pipeline called Magicmol; this pipeline has a re-mastered RNN network and utilize SELFIES presentation instead of SMILES. Our backbone model achieved extraordinary performance while reducing the training cost; moreover, we devised reward truncate strategies to eliminate the model collapse problem. Additionally, adopting SELFIES presentation made it possible to combine STONED-SELFIES as a post-processing procedure for specified molecule optimization and quick chemical space exploration.

Novel Molecular Representations Using Neumann-Cayley Orthogonal Gated Recurrent Unit

Advances in deep neural networks (DNNs) have made a very powerful machine learning method available to researchers across many fields of study, including the biomedical and cheminformatics communities, where DNNs help to improve tasks such as protein performance, molecular design, drug discovery, etc. Many of those tasks rely on molecular descriptors for representing molecular characteristics in cheminformatics. Despite significant efforts and the introduction of numerous methods that derive molecular descriptors, the quantitative prediction of molecular properties remains challenging. One widely used method of encoding molecule features into bit strings is the molecular fingerprint. In this work, we propose using new Neumann-Cayley Gated Recurrent Units (NC-GRU) inside the Neural Nets encoder (AutoEncoder) to create neural molecular fingerprints (NC-GRU fingerprints). The NC-GRU AutoEncoder introduces orthogonal weights into widely used GRU architecture, resulting in faster, more stable training, and more reliable molecular fingerprints. Integrating novel NC-GRU fingerprints and Multi-Task DNN schematics improves the performance of various molecular-related tasks such as toxicity, partition coefficient, lipophilicity, and solvation-free energy, producing state-of-the-art results on several benchmarks.

Discovery of Potential Inhibitors of SARS-CoV-2 Main Protease by a Transfer Learning Method

The COVID-19 pandemic caused by SARS-CoV-2 remains a global public health threat and has prompted the development of antiviral therapies. Artificial intelligence may be one of the strategies to facilitate drug development for emerging and re-emerging diseases. The main protease (Mpro) of SARS-CoV-2 is an attractive drug target due to its essential role in the virus life cycle and high conservation among SARS-CoVs. In this study, we used a data augmentation method to boost transfer learning model performance in screening for potential inhibitors of SARS-CoV-2 Mpro. This method appeared to outperform graph convolution neural network, random forest and Chemprop on an external test set. The fine-tuned model was used to screen for a natural compound library and a de novo generated compound library. By combination with other in silico analysis methods, a total of 27 compounds were selected for experimental validation of anti-Mpro activities. Among all the selected hits, two compounds (gyssypol acetic acid and hyperoside) displayed inhibitory effects against Mpro with IC50 values of 67.6 μM and 235.8 μM, respectively. The results obtained in this study may suggest an effective strategy of discovering potential therapeutic leads for SARS-CoV-2 and other coronaviruses.

Characterizing the ligand-binding affinity toward SARS-CoV-2 Mpro via physics- and knowledge-based approaches

Computational approaches, including physics- and knowledge-based methods, have commonly been used to determine the ligand-binding affinity toward SARS-CoV-2 main protease (Mpro or 3CLpro). Strong binding ligands can thus be suggested as potential inhibitors for blocking the biological activity of the protease. In this context, this paper aims to provide a short review of computational approaches that have recently been applied in the search for inhibitor candidates of Mpro. In particular, molecular docking and molecular dynamics (MD) simulations are usually combined to predict the binding affinity of thousands of compounds. Quantitative structure-activity relationship (QSAR) is the least computationally demanding and therefore can be used for large chemical collections of ligands. However, its accuracy may not be high. Moreover, the quantum mechanics/molecular mechanics (QM/MM) method is most commonly used for covalently binding inhibitors, which also play an important role in inhibiting the activity of SARS-CoV-2. Furthermore, machine learning (ML) models can significantly increase the searching space of ligands with high accuracy for binding affinity prediction. Physical insights into the binding process can then be confirmed via physics-based calculations. Integration of ML models into computational chemistry provides many more benefits and can lead to new therapies sooner.

ReMODE: a deep learning-based web server for target-specific drug design

Deep learning (DL) and machine learning contribute significantly to basic biology research and drug discovery in the past few decades. Recent advances in DL-based generative models have led to superior developments in de novo drug design. However, data availability, deep data processing, and the lack of user-friendly DL tools and interfaces make it difficult to apply these DL techniques to drug design. We hereby present ReMODE (Receptor-based MOlecular DEsign), a new web server based on DL algorithm for target-specific ligand design, which integrates different functional modules to enable users to develop customizable drug design tasks. As designed, the ReMODE sever can construct the target-specific tasks toward the protein targets selected by users. Meanwhile, the server also provides some extensions: users can optimize the drug-likeness or synthetic accessibility of the generated molecules, and control other physicochemical properties; users can also choose a sub-structure/scaffold as a starting point for fragment-based drug design. The ReMODE server also enables users to optimize the pharmacophore matching and docking conformations of the generated molecules. We believe that the ReMODE server will benefit researchers for drug discovery. ReMODE is publicly available at http://cadd.zju.edu.cn/relation/remode/ .

Exploration of Chemical Space Guided by PixelCNN for Fragment-Based De Novo Drug Discovery

We report a novel framework for achieving fragment-based molecular design using pixel convolutional neural network (PixelCNN) combined with the simplified molecular input line entry system (SMILES) as molecular representation. While a widely used recurrent neural network (RNN) assumes monotonically decaying correlations in strings, PixelCNN captures a periodicity among characters of SMILES. Thus, PixelCNN provides us with a novel solution for the analysis of chemical space by extracting the periodicity of molecular structures that will be buried in SMILES. Moreover, this characteristic enables us to generate molecules by combining several simple building blocks, such as a benzene ring and side-chain structures, which contributes to the effective exploration of chemical space by step-by-step searching for molecules from a target fragment. In conclusion, PixelCNN could be a powerful approach focusing on the periodicity of molecules to explore chemical space for the fragment-based molecular design.

Proteolysis and Deficiency of α1-Proteinase Inhibitor in SARS-CoV-2 Infection

The SARS-CoV-2 pandemic had stimulated the emergence of numerous publications on the α1-proteinase inhibitor (α1-PI, α1-antitrypsin), especially when it was found that the regions of high mortality corresponded to the regions with deficient α1-PI alleles. By analogy with the data obtained in the last century, when the first cause of the genetic deficiency of α1-antitrypsin leading to elastase activation in pulmonary emphysema was proven, it can be supposed that proteolysis hyperactivation in COVID-19 may be associated with the impaired functions of α1-PI. The purpose of this review was to systematize the scientific data and critical directions for translational studies on the role of α1-PI in SARS-CoV-2-induced proteolysis hyperactivation as a diagnostic marker and a therapeutic target. This review describes the proteinase-dependent stages of viral infection: the reception and penetration of the virus into a cell and the imbalance of the plasma aldosterone-angiotensin-renin, kinin, and blood clotting systems. The role of ACE2, TMPRSS, ADAM17, furin, cathepsins, trypsin- and elastase-like serine proteinases in the virus tropism, the activation of proteolytic cascades in blood, and the COVID-19-dependent complications is considered. The scientific reports on α1-PI involvement in the SARS-CoV-2-induced inflammation, the relationship with the severity of infection and comorbidities were analyzed. Particular attention is paid to the acquired α1-PI deficiency in assessing the state of patients with proteolysis overactivation and chronic non-inflammatory diseases, which are accompanied by the risk factors for comorbidity progression and the long-term consequences of COVID-19. Essential data on the search and application of protease inhibitor drugs in the therapy for bronchopulmonary and cardiovascular pathologies were analyzed. The evidence of antiviral, anti-inflammatory, anticoagulant, and anti-apoptotic effects of α1-PI, as well as the prominent data and prospects for its application as a targeted drug in the SARS-CoV-2 acquired pneumonia and related disorders, are presented.

Conditional reduction of the loss value versus reinforcement learning for biassing a de-novo drug design generator

Deep learning has demonstrated promising results in de novo drug design. Often, the general pipeline consists of training a generative model (G) to learn the building rules of valid molecules, then using a biassing technique such as reinforcement learning (RL) to focus G on the desired chemical space. However, this sequential training of the same model for different tasks is known to be prone to a catastrophic forgetting (CF) phenomenon. This work presents a novel yet simple approach to bias G with significantly less CF than RL. The proposed method relies on backpropagating a reduced value of the cross-entropy loss used to train G according to the proportion of desired molecules that the biased-G can generate. We named our approach CRLV, short for conditional reduction of the loss value. We compared the two biased models (RL-biased-G and CRLV-biased-G) for four different objectives related to de novo drug design.CRLV-biased-G outperformed RL-biased-G in all four objectives and manifested appreciably less CF. Besides, an intersection analysis between molecules generated by the RL-biased-G and the CRLV-biased-G revealed that they can be used jointly without losing diversity given the low percentage of overlap between the two to further increase the desirability. Finally, we show that the difficulty of an objective is proportional to (i) its frequency in the dataset used to train G and (ii) the associated structural variance (SV), which is a new parameter we introduced in this paper, calling for novel exploration techniques for such difficult objectives.

A Transfer Learning Approach for Reaction Discovery in Small Data Situations Using Generative Model

Sustainable practices in chemical sciences can be better realized by adopting interdisciplinary approaches that combine the advantages of machine learning (ML) on the initially acquired small data in reaction discovery. Developing new reactions generally remains heuristic and even time and resource intensive. For instance, synthesis of fluorine-containing compounds, which constitute ∼20% of the marketed drugs, relies on deoxyfluorination of abundantly available alcohols. Herein, we demonstrate the use of a recurrent neural network-based deep generative model built on a library of just 37 alcohols for effective learning and exploration of the chemical space. The proof-of-concept ML model is able to generate good quality, synthetically accessible, higher-yielding novel alcohol molecules. This protocol would have superior utility for deployment into a practical reaction discovery pipeline.

•

Dual pronged transfer learning, both to generate and predict yields of new molecules

•

Demonstrated the utility for an important family of deoxyfluorination of alcohols

•

Applicable for practically more likely situations with relatively smaller data

•

Extendable to other reaction manifolds to facilitate expedited reaction discovery

[Proteolysis and deficiency of α1-proteinase inhibitor in SARS-CoV-2 infection]

The SARS-CoV-2 pandemia had stimulated the numerous publications emergence on the α1-proteinase inhibitor (α1-PI, α1-antitrypsin), primarily when it was found that high mortality in some regions corresponded to the regions with deficient α1-PI alleles. By analogy with the last century's data, when the root cause of the α1-antitrypsin, genetic deficiency leading to the elastase activation in pulmonary emphysema, was proven. It is evident that proteolysis hyperactivation in COVID-19 may be associated with α1-PI impaired functions. The purpose of this review is to systematize scientific data, critical directions for translational studies on the role of α1-PI in SARS-CoV-2-induced proteolysis hyperactivation as a diagnostic marker and a target in therapy. This review describes the proteinase-dependent stages of a viral infection: the reception and virus penetration into the cell, the plasma aldosterone-angiotensin-renin, kinins, blood clotting systems imbalance. The ACE2, TMPRSS, ADAM17, furin, cathepsins, trypsin- and elastase-like serine proteinases role in the virus tropism, proteolytic cascades activation in blood, and the COVID-19-dependent complications is presented. The analysis of scientific reports on the α1-PI implementation in the SARS-CoV-2-induced inflammation, the links with the infection severity, and comorbidities were carried out. Particular attention is paid to the acquired α1-PI deficiency in assessing the patients with the proteolysis overactivation and chronic non-inflammatory diseases that are accompanied by the risk factors for the comorbidities progression, and the long-term consequences of COVID-19 initiation. Analyzed data on the search and proteases inhibitory drugs usage in the bronchopulmonary cardiovascular pathologies therapy are essential. It becomes evident the antiviral, anti-inflammatory, anticoagulant, anti-apoptotic effect of α1-PI. The prominent data and prospects for its application as a targeted drug in the SARS-CoV-2 acquired pneumonia and related disorders are presented.

Approaches for enhancing the analysis of chemical space for drug discovery

INTRODUCTION

Chemical space is a powerful, general, and practical conceptual framework in drug discovery and other areas in chemistry that addresses the diversity of molecules and it has various applications. Moreover, chemical space is a cornerstone of chemoinformatics as a scientific discipline. In response to the increase in the set of chemical compounds in databases, generators of chemical structures, and tools to calculate molecular descriptors, novel approaches to generate visual representations of chemical space in low dimensions are emerging and evolving. Such approaches include a wide range of commercial and free applications, software, and open-source methods.

AREAS COVERED

The current state of chemical space in drug design and discovery is reviewed. The topics discussed herein include advances for efficient navigation in chemical space, the use of this concept in assessing the diversity of different data sets, exploring structure-property/activity relationships for one or multiple endpoints, and compound library design. Recent advances in methodologies for generating visual representations of chemical space have been highlighted, thereby emphasizing open-source methods.

EXPERT OPINION

Quantitative and qualitative generation and analysis of chemical space require novel approaches for handling the increasing number of molecules and their information available in chemical databases (including emerging ultra-large libraries). In addition, it is of utmost importance to note that chemical space is a conceptual framework that goes beyond visual representation in low dimensions. However, the graphical representation of chemical space has several practical applications in drug discovery and beyond.

Machine learning assessment of the binding region as a tool for more efficient computational receptor-ligand docking

We present a combined computational approach to protein-ligand binding, which consists of two steps: (1) a deep neural network is used to locate a binding region on a target protein, and (2) molecular docking of a ligand is performed within the specified region to obtain the best pose using Autodock Vina. Our in-house designed neural network was trained using the PepBDB dataset. Although the training dataset consisted of protein-peptide complexes, we show that the approach is not limited to peptides, but also works remarkably well for a large class of non-peptide ligands. The results are compared with those in which the binding region (first step) was provided by Accluster. In cases where no prior experimental data on the binding region are available, our deep neural network provides a fast and effective alternative to classical software for its localization. Our code is available at https://github.com/mksmd/NNforDocking.

How can natural language processing help model informed drug development?: a review

Objective

To summarize applications of natural language processing (NLP) in model informed drug development (MIDD) and identify potential areas of improvement.

Materials and Methods

Publications found on PubMed and Google Scholar, websites and GitHub repositories for NLP libraries and models. Publications describing applications of NLP in MIDD were reviewed. The applications were stratified into 3 stages: drug discovery, clinical trials, and pharmacovigilance. Key NLP functionalities used for these applications were assessed. Programming libraries and open-source resources for the implementation of NLP functionalities in MIDD were identified.

Results

NLP has been utilized to aid various processes in drug development lifecycle such as gene-disease mapping, biomarker discovery, patient-trial matching, adverse drug events detection, etc. These applications commonly use NLP functionalities of named entity recognition, word embeddings, entity resolution, assertion status detection, relation extraction, and topic modeling. The current state-of-the-art for implementing these functionalities in MIDD applications are transformer models that utilize transfer learning for enhanced performance. Various libraries in python, R, and Java like huggingface, sparkNLP, and KoRpus as well as open-source platforms such as DisGeNet, DeepEnroll, and Transmol have enabled convenient implementation of NLP models to MIDD applications.

Discussion

Challenges such as reproducibility, explainability, fairness, limited data, limited language-support, and security need to be overcome to ensure wider adoption of NLP in MIDD landscape. There are opportunities to improve the performance of existing models and expand the use of NLP in newer areas of MIDD.

Conclusions

This review provides an overview of the potential and pitfalls of current NLP approaches in MIDD.

Artificial Intelligence Technologies for COVID-19 De Novo Drug Design

The recent covid crisis has provided important lessons for academia and industry regarding digital reorganization. Among the fascinating lessons from these times is the huge potential of data analytics and artificial intelligence. The crisis exponentially accelerated the adoption of analytics and artificial intelligence, and this momentum is predicted to continue into the 2020s and beyond. Drug development is a costly and time-consuming business, and only a minority of approved drugs generate returns exceeding the research and development costs. As a result, there is a huge drive to make drug discovery cheaper and faster. With modern algorithms and hardware, it is not too surprising that the new technologies of artificial intelligence and other computational simulation tools can help drug developers. In only two years of covid research, many novel molecules have been designed/identified using artificial intelligence methods with astonishing results in terms of time and effectiveness. This paper reviews the most significant research on artificial intelligence in de novo drug design for COVID-19 pharmaceutical research.

Generative machine learning for de novo drug discovery: A systematic review

Recent research on artificial intelligence indicates that machine learning algorithms can auto-generate novel drug-like molecules. Generative models have revolutionized de novo drug discovery, rendering the explorative process more efficient. Several model frameworks and input formats have been proposed to enhance the performance of intelligent algorithms in generative molecular design. In this systematic literature review of experimental articles and reviews over the last five years, machine learning models, challenges associated with computational molecule design along with proposed solutions, and molecular encoding methods are discussed. A query-based search of the PubMed, ScienceDirect, Springer, Wiley Online Library, arXiv, MDPI, bioRxiv, and IEEE Xplore databases yielded 87 studies. Twelve additional studies were identified via citation searching. Of the articles in which machine learning was implemented, six prominent algorithms were identified: long short-term memory recurrent neural networks (LSTM-RNNs), variational autoencoders (VAEs), generative adversarial networks (GANs), adversarial autoencoders (AAEs), evolutionary algorithms, and gated recurrent unit (GRU-RNNs). Furthermore, eight central challenges were designated: homogeneity of generated molecular libraries, deficient synthesizability, limited assay data, model interpretability, incapacity for multi-property optimization, incomparability, restricted molecule size, and uncertainty in model evaluation. Molecules were encoded either as strings, which were occasionally augmented using randomization, as 2D graphs, or as 3D graphs. Statistical analysis and visualization are performed to illustrate how approaches to machine learning in de novo drug design have evolved over the past five years. Finally, future opportunities and reservations are discussed.

Complex machine learning model needs complex testing: Examining predictability of molecular binding affinity by a graph neural network

Drug discovery pipelines typically involve high-throughput screening of large amounts of compounds in a search of potential drugs candidates. As a chemical space of small organic molecules is huge, a "navigation" over it urges for fast and lightweight computational methods, thus promoting machine-learning approaches for processing huge pools of candidates. In this contribution, we present a graph-based deep neural network for prediction of protein-drug binding affinity and assess its predictive power under thorough testing conditions. Within the suggested approach, both protein and drug molecules are represented as graphs and passed to separate graph sub-networks, then concatenated and regressed towards a binding affinity. The neural network is trained on two binding affinity datasets-PDBbind and data imported from RCSB Protein Data Bank. In order to explore the generalization capabilities of the model we go beyond traditional random or leave-cluster-out techniques and demonstrate the need for more elaborate model performance assessment - six different strategies for test/train data partitioning (random, time- and property-arranged, protein- and ligand-clustered) with a k-fold cross-validation are engaged. Finally, we discuss the model performance in terms of a set of metrics for different split strategies and fold arrangement. Our code is available at https://github.com/SoftServeInc/affinity-by-GNN.

A highly accurate metadynamics-based Dissociation Free Energy method to calculate protein-protein and protein-ligand binding potencies

Although seeking to develop a general and accurate binding free energy calculation method for protein-protein and protein-ligand interactions has been a continuous effort for decades, only limited successes have been obtained so far. Here, we report the development of a metadynamics-based procedure that calculates Dissociation Free Energy (DFE) and its application to 19 non-congeneric protein-protein complexes and hundreds of protein-ligand complexes covering eight targets. We achieved very high correlations in comparison to experimental binding free energies for these diverse sets of systems, demonstrating the generality and accuracy of the method. Since structures of most proteins are available owing to the recent success of prediction by artificial intelligence, a general free energy method such as DFE, combined with other methods, can make structure-based drug design a widely viable and reliable solution to develop both traditional small molecule drugs and biologic drugs as well as PROTACS.

Machine learning prediction of 3CLpro SARS-CoV-2 docking scores

Molecular docking results of two training sets containing 866 and 8,696 compounds were used to train three different machine learning (ML) approaches. Neural network approaches according to Keras and TensorFlow libraries and the gradient boosted decision trees approach of XGBoost were used with DScribe’s Smooth Overlap of Atomic Positions molecular descriptors. In addition, neural networks using the SchNetPack library and descriptors were used. The ML performance was tested on three different sets, including compounds for future organic synthesis. The final evaluation of the ML predicted docking scores was based on the ZINC in vivo set, from which 1,200 compounds were randomly selected with respect to their size. The results obtained showed a consistent ML prediction capability of docking scores, and even though compounds with more than 60 atoms were found slightly overestimated they remain valid for a subsequent evaluation of their drug repurposing suitability.

Binding mechanism of inhibitors to SARS-CoV-2 main protease deciphered by multiple replica molecular dynamics simulations

The outbreak caused by SARS-CoV-2 has received extensive worldwide attention. As the main protease (M^pro) in SARS-CoV-2 has no human homologues, it is feasible to reduce the possibility of targeting the host protein by accidental drugs. Thus, M^pro has been an attractive target of efficient drug design for anti-SARS-CoV-2 treatment. In this work, multiple replica molecular dynamics (MRMD) simulations, principal component analysis (PCA), free energy landscapes (FELs), and the molecular mechanics-generalized Born surface area (MM-GBSA) method were integrated together to decipher the binding mechanism of four inhibitors masitinib, O6K, FJC and GQU to M^pro. The results indicate that the binding of four inhibitors clearly affects the structural flexibility and internal dynamics of M^pro along with dihedral angle changes of key residues. The analysis of FELs unveils that the stability in the relative orientation and geometric position of inhibitors to M^pro is favorable for inhibitor binding. Residue-based free energy decomposition reveals that the inhibitor-M^pro interaction networks involving hydrogen bonding interactions and hydrophobic interactions provide significant information for the design of potent inhibitors against M^pro. The hot spot residues including H41, M49, F140, N142, G143, C145, H163, H164, M165, E166 and Q189 identified by computational alanine scanning are considered as reliable targets of clinically available inhibitors inhibiting the activities of M^pro.

Artificial intelligence methods to repurpose and discover new drugs to fight the Coronavirus disease-2019 pandemic

The Coronavirus disease 2019 pandemic struck the world at the end of 2019 and, as of 2021, there are no specific drugs available against the causative agent, the severe acute respiratory syndrome-Coronavirus-2 (SARS-CoV-2). From the onset of the pandemic, researchers have been trying to find drugs among the current therapeutic arsenal that could target crucial viral function, and many of these efforts resulted in clinical trials to repurpose a drug for this new indication. In this scenario, artificial intelligence (AI) is of fundamental importance, allowing academia and pharmaceutical companies to accelerate the discovery of biochemical insights from the chemical and biological information available in literature databases. This chapter will cover some AI methods that are being explored to repurpose drugs against SARS-CoV-2. It will be outlined how these methods work followed by a discussion of selected examples applying them to identify promising drugs.

Identification of Potential Ligands of the Main Protease of Coronavirus SARS-CoV-2 (2019-nCoV) Using Multimodal Generative Neural-Networks

The recent outbreak of coronavirus disease 2019 (COVID-19) is posing a global threat to human population. The pandemic caused by novel coronavirus (2019-nCoV), also called as severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2); first emerged in Wuhan city, Hubei province of China in December 2019. The rapid human to human transmission has caused the contagion to spread world-wide affecting 244,385,444 (244.4 million) people globally causing 4,961,489 (5 million) fatalities dated by 27 October 2021. At present, 6,697,607,393 (6.7 billion) vaccine doses have been administered dated by 27 October 2021, for the prevention of COVID-19 infections. Even so, this critical and threatening situation of pandemic and due to various variants’ emergence, the pandemic control has become challenging; this calls for gigantic efforts to find new potent drug candidates and effective therapeutic approaches against the virulent respiratory disease of COVID-19. In the respiratory morbidities of COVID-19, the functionally crucial drug target for the antiviral treatment could be the main protease/3-chymotrypsin protease (Mpro/3CLpro) enzyme that is primarily involved in viral maturation and replication. In view of this, in the current study I have designed a library of small molecules against the main protease (Mpro) of coronavirus SARS-CoV-2 (2019-nCoV) by using multimodal generative neural-networks. The scaffold-based molecular docking of the series of compounds at the active site of the protein was performed; binding poses of the molecules were evaluated and protein-ligand interaction studies followed by the binding affinity calculations validated the findings. I have identified a number of small promising lead compounds that could serve as potential inhibitors of the main protease (Mpro) enzyme of coronavirus SARS-CoV-2 (2019-nCoV). This study would serve as a step forward in the development of effective antiviral therapeutic agents against the COVID-19.

Comprehensive assessment of deep generative architectures for de novo drug design

Recently, deep learning (DL)-based de novo drug design represents a new trend in pharmaceutical research, and numerous DL-based methods have been developed for the generation of novel compounds with desired properties. However, a comprehensive understanding of the advantages and disadvantages of these methods is still lacking. In this study, the performances of different generative models were evaluated by analyzing the properties of the generated molecules in different scenarios, such as goal-directed (rediscovery, optimization and scaffold hopping of active compounds) and target-specific (generation of novel compounds for a given target) tasks. In overall, the DL-based models have significant advantages over the baseline models built by the traditional methods in learning the physicochemical property distributions of the training sets and may be more suitable for target-specific tasks. However, both the baselines and DL-based generative models cannot fully exploit the scaffolds of the training sets, and the molecules generated by the DL-based methods even have lower scaffold diversity than those generated by the traditional models. Moreover, our assessment illustrates that the DL-based methods do not exhibit obvious advantages over the genetic algorithm-based baselines in goal-directed tasks. We believe that our study provides valuable guidance for the effective use of generative models in de novo drug design.

De novo design of novel protease inhibitor candidates in the treatment of SARS-CoV-2 using deep learning, docking, and molecular dynamic simulations

The main protease of SARS-CoV-2 is a critical target for the design and development of antiviral drugs. 2.5 M compounds were used in this study to train an LSTM generative network via transfer learning in order to identify the four best candidates capable of inhibiting the main proteases in SARS-CoV-2. The network was fine-tuned over ten generations, with each generation resulting in higher binding affinity scores. The binding affinities and interactions between the selected candidates and the SARS-CoV-2 main protease are predicted using a molecular docking simulation using AutoDock Vina. The compounds selected have a strong interaction with the key MET 165 and Cys145 residues. Molecular dynamics (MD) simulations were run for 150ns to validate the docking results on the top four ligands. Additionally, root-mean-square deviation (RMSD), root-mean-square fluctuation (RMSF), and hydrogen bond analysis strongly support these findings. Furthermore, the MM-PBSA free energy calculations revealed that these chemical molecules have stable and favorable energies, resulting in a strong binding with Mpro's binding site. This study's extensive computational and statistical analyses indicate that the selected candidates may be used as potential inhibitors against the SARS-CoV-2 in-silico environment. However, additional in-vitro, in-vivo, and clinical trials are required to demonstrate their true efficacy.

Virus structure and structure-based antivirals

Structure-based antiviral developments in the past two years have been dominated by the structure determination and inhibition of SARS-CoV-2 proteins and new lead molecules for picornaviruses. The SARS-CoV-2 spike protein has been targeted successfully with antibodies, nanobodies, and receptor protein mimics effectively blocking receptor binding or fusion. The two most promising non-structural proteins sharing strong structural and functional conservation across virus families are the main protease and the RNA-dependent RNA polymerase, for which design and reuse of broad range inhibitors already approved for use has been an attractive avenue. For picornaviruses, the increasing recognition of the transient expansion of the capsid as a critical transition towards RNA release has been targeted through a newly identified, apparently widely conserved, druggable, interprotomer pocket preventing viral entry. We summarize some of the key papers in these areas and ponder the practical uses and contributions of molecular modeling alongside empirical structure determination.

Structure-guided design of a perampanel-derived pharmacophore targeting the SARS-CoV-2 main protease

There is a clinical need for direct-acting antivirals targeting SARS-CoV-2, the coronavirus responsible for the COVID-19 pandemic, to complement current therapeutic strategies. The main protease (Mpro) is an attractive target for antiviral therapy. However, the vast majority of protease inhibitors described thus far are peptidomimetic and bind to the active-site cysteine via a covalent adduct, which is generally pharmacokinetically unfavorable. We have reported the optimization of an existing FDA-approved chemical scaffold, perampanel, to bind to and inhibit Mpro noncovalently with IC50s in the low-nanomolar range and EC50s in the low-micromolar range. Here, we present nine crystal structures of Mpro bound to a series of perampanel analogs, providing detailed structural insights into their mechanism of action and structure-activity relationship. These insights further reveal strategies for pursuing rational inhibitor design efforts in the context of considerable active-site flexibility and potential resistance mechanisms.

ADVIAN: Alzheimer's Disease VGG-Inspired Attention Network Based on Convolutional Block Attention Module and Multiple Way Data Augmentation

Aim: Alzheimer's disease is a neurodegenerative disease that causes 60-70% of all cases of dementia. This study is to provide a novel method that can identify AD more accurately. Methods: We first propose a VGG-inspired network (VIN) as the backbone network and investigate the use of attention mechanisms. We proposed an Alzheimer's Disease VGG-Inspired Attention Network (ADVIAN), where we integrate convolutional block attention modules on a VIN backbone. Also, 18-way data augmentation is proposed to avoid overfitting. Ten runs of 10-fold cross-validation are carried out to report the unbiased performance. Results: The sensitivity and specificity reach 97.65 ± 1.36 and 97.86 ± 1.55, respectively. Its precision and accuracy are 97.87 ± 1.53 and 97.76 ± 1.13, respectively. The F1 score, MCC, and FMI are obtained as 97.75 ± 1.13, 95.53 ± 2.27, and 97.76 ± 1.13, respectively. The AUC is 0.9852. Conclusion: The proposed ADVIAN gives better results than 11 state-of-the-art methods. Besides, experimental results demonstrate the effectiveness of 18-way data augmentation.

Ensembling machine learning models to boost molecular affinity prediction

This study unites six popular machine learning approaches to enhance the prediction of a molecular binding affinity between receptors (large protein molecules) and ligands (small organic molecules). Here we examine a scheme where affinity of ligands is predicted against a single receptor - human thrombin, thus, the models consider ligand features only. However, the suggested approach can be repurposed for other receptors. The methods include Support Vector Machine, Random Forest, CatBoost, feed-forward neural network, graph neural network, and Bidirectional Encoder Representations from Transformers. The first five methods use input features based on physico-chemical properties of molecules, while the last one is based on textual molecular representations. All approaches do not rely on atomic spatial coordinates, avoiding a potential bias from known structures, and are capable of generalizing for compounds with unknown conformations. Within each of the methods, we have trained two models that solve classification and regression tasks. Then, all models are grouped into a pipeline of two subsequent ensembles. The first ensemble aggregates six classification models which vote whether a ligand binds to a receptor or not. If a ligand is classified as active (i.e., binds), the second ensemble predicts its binding affinity in terms of the inhibition constant K_i.