Active Search for Computer-aided Drug Design

Recent Advances in Automated Structure-Based De Novo Drug Design

As the number of determined and predicted protein structures and the size of druglike 'make-on-demand' libraries soar, the time-consuming nature of structure-based computer-aided drug design calls for innovative computational algorithms. De novo drug design introduces in silico heuristics to accelerate searching in the vast chemical space. This review focuses on recent advances in structure-based de novo drug design, ranging from conventional fragment-based methods, evolutionary algorithms, and Metropolis Monte Carlo methods to deep generative models. Due to the historical limitation of de novo drug design generating readily available drug-like molecules, we highlight the synthetic accessibility efforts in each category and the benchmarking strategies taken to validate the proposed framework.

Active learning of enhancer and silencer regulatory grammar in photoreceptors

Cis-regulatory elements (CREs) direct gene expression in health and disease, and models that can accurately predict their activities from DNA sequences are crucial for biomedicine. Deep learning represents one emerging strategy to model the regulatory grammar that relates CRE sequence to function. However, these models require training data on a scale that exceeds the number of CREs in the genome. We address this problem using active machine learning to iteratively train models on multiple rounds of synthetic DNA sequences assayed in live mammalian retinas. During each round of training the model actively selects sequence perturbations to assay, thereby efficiently generating informative training data. We iteratively trained a model that predicts the activities of sequences containing binding motifs for the photoreceptor transcription factor Cone-rod homeobox (CRX) using an order of magnitude less training data than current approaches. The model's internal confidence estimates of its predictions are reliable guides for designing sequences with high activity. The model correctly identified critical sequence differences between active and inactive sequences with nearly identical transcription factor binding sites, and revealed order and spacing preferences for combinations of motifs. Our results establish active learning as an effective method to train accurate deep learning models of cis-regulatory function after exhausting naturally occurring training examples in the genome.

Molecular Modeling Insights into the Structure and Behavior of Integrins: A Review

Integrins are heterodimeric glycoproteins crucial to the physiology and pathology of many biological functions. As adhesion molecules, they mediate immune cell trafficking, migration, and immunological synapse formation during inflammation and cancer. The recognition of the vital roles of integrins in various diseases revealed their therapeutic potential. Despite the great effort in the last thirty years, up to now, only seven integrin-based drugs have entered the market. Recent progress in deciphering integrin functions, signaling, and interactions with ligands, along with advancement in rational drug design strategies, provide an opportunity to exploit their therapeutic potential and discover novel agents. This review will discuss the molecular modeling methods used in determining integrins' dynamic properties and in providing information toward understanding their properties and function at the atomic level. Then, we will survey the relevant contributions and the current understanding of integrin structure, activation, the binding of essential ligands, and the role of molecular modeling methods in the rational design of antagonists. We will emphasize the role played by molecular modeling methods in progress in these areas and the designing of integrin antagonists.

TNF-α Inhibitors from Natural Compounds: An Overview, CADD Approaches, and their Exploration for Anti-inflammatory Agents

Inflammation is a natural process that occurs in the organism in response to harmful external agents. Despite being considered beneficial, exaggerated cases can cause severe problems for the body. The main inflammatory manifestations are pain, increased temperature, edema, decreased mobility, and quality of life for affected individuals. Diseases such as arthritis, cancer, allergies, infections, arteriosclerosis, neurodegenerative diseases, and metabolic problems are mainly characterized by an exaggerated inflammatory response. Inflammation is related to two categories of substances: pro- and anti-inflammatory mediators. Among the pro-inflammatory mediators is Tumor Necrosis Factor-α (TNF-α). It is associated with immune diseases, cancer, and psychiatric disorders which increase its excretion. Thus, it becomes a target widely used in discovering new antiinflammatory drugs. In this context, secondary metabolites biosynthesized by plants have been used for thousands of years and continue to be one of the primary sources of new drug scaffolds against inflammatory diseases. To decrease costs related to the drug discovery process, Computer-Aided Drug Design (CADD) techniques are broadly explored to increase the chances of success. In this review, the main natural compounds derived from alkaloids, flavonoids, terpene, and polyphenols as promising TNF-α inhibitors will be discussed. Finally, we applied a molecular modeling protocol involving all compounds described here, suggesting that their interactions with Tyr⁵⁹, Tyr¹¹⁹, Tyr¹⁵¹, Leu⁵⁷, and Gly¹²¹ residues are essential for the activity. Such findings can be useful for research groups worldwide to design new anti-inflammatory TNF-α inhibitors.

Active Knowledge Extraction from Cyclic Voltammetry

Cyclic Voltammetry (CV) is an electro-chemical characterization technique used in an initial material screening for desired properties and to extract information about electro-chemical reactions. In some applications, to extract kinetic information of the associated reactions (e.g., rate constants and turn over frequencies), CV curve should have a specific shape (for example an S-shape). However, often the characterization settings to obtain such curve are not known a priori. In this paper, an active search framework is defined to accelerate identification of characterization settings that enable knowledge extraction from CV experiments. Towards this goal, a representation of CV responses is used in combination with Bayesian Model Selection (BMS) method to efficiently label the response to be either S-shape or not S-shape. Using an active search with BMS oracle, we report a linear target identification in a six-dimensional search space (comprised of thermodynamic, mass transfer, and solution variables as dimensions). Our framework has the potential to be a powerful virtual screening technique for molecular catalysts, bi-functional fuel cell catalysts, and other energy conversion and storage systems.

Anti-Inflammatory Effects of Ginsenoside Rb3 in LPS-Induced Macrophages Through Direct Inhibition of TLR4 Signaling Pathway

Panax ginseng has therapeutic effects on various inflammation-related diseases. Ginsenoside Rb3 (GRb3), a natural compound with anti-inflammatory and immunomodulatory properties, is one of the main active panaxadiol extracted from Panax ginseng. We explored whether GRb3 inhibited LPS-mediated inflammation through TLR4/NF-κB/MAPK signaling in macrophages. GRb3 attenuated NO and PGE2 production by attenuating iNOS and COX2 expression. GRb3 also suppressed pro-inflammatory cytokines levels, including IL-1β, IL-6, and TNF-α. Moreover, GRb3 administration significantly suppressed NF-κB (p65) nuclear translocation and the phosphorylation levels of p65, IκBα, JNK, p38, and ERK dose-dependently. Molecular docking demonstrated that GRb3 could dock onto the hydrophobic binding site of TLR4/MD2 complex, with a binding energy of −8.79 kcal/mol. Molecular dynamics (MD) displayed stable TLR4-MD2-GRb3 modeling. GRb3 dose-dependently inhibited LPS binding to cell membranes and blocked TLR4 expression. Surface plasmon resonance imaging (SPRi) revealed that GRb3 had an excellent binding affinity to TLR4/MD2 complex. Notably, resatorvid (TAK242), a selective TLR4 inhibitor, did not increase the repressive influence of GRb3 in RAW264.7 macrophages. Moreover, TLR4 overexpression partially reversed the repressive roles of GRb3 on the NF-κB/MAPK pathway and inflammatory mediators. Collectively, our study strongly indicated that GRb3 attenuated LPS-mediated inflammation through direct inhibition of TLR4 signaling. A novel insight into the underlying mechanism of anti-inflammatory effects of GRb3 in macrophages was confirmed.

Molecular Simulation of αvβ6 Integrin Inhibitors

The urgent need for new treatments for the chronic lung disease idiopathic pulmonary fibrosis (IPF) motivates research into antagonists of the RGD binding integrin αvβ6, a protein linked to the initiation and progression of the disease. Molecular dynamics (MD) simulations of αvβ6 in complex with its natural ligand, pro-TGF-β1, show the persistence over time of a bidentate Arg-Asp ligand-receptor interaction and a metal chelate interaction between an aspartate on the ligand and an Mg²⁺ ion in the active site. This is typical of RGD binding ligands. Additional binding site interactions, which are not observed in the static crystal structure, are also identified. We investigate an RGD mimetic, which serves as a framework for a series of potential αvβ6 antagonists. The scaffold includes a derivative of the widely utilized 1,8-naphthyridine moiety, for which we present force field parameters, to enable MD and relative free energy perturbation (FEP) simulations. The MD simulations highlight the importance of hydrogen bonding and cation-π interactions. The FEP calculations predict relative binding affinities, within 1.5 kcal mol^-1, on average, of experiments.

Structure-Based Virtual Screening: From Classical to Artificial Intelligence

The drug development process is a major challenge in the pharmaceutical industry since it takes a substantial amount of time and money to move through all the phases of developing of a new drug. One extensively used method to minimize the cost and time for the drug development process is computer-aided drug design (CADD). CADD allows better focusing on experiments, which can reduce the time and cost involved in researching new drugs. In this context, structure-based virtual screening (SBVS) is robust and useful and is one of the most promising in silico techniques for drug design. SBVS attempts to predict the best interaction mode between two molecules to form a stable complex, and it uses scoring functions to estimate the force of non-covalent interactions between a ligand and molecular target. Thus, scoring functions are the main reason for the success or failure of SBVS software. Many software programs are used to perform SBVS, and since they use different algorithms, it is possible to obtain different results from different software using the same input. In the last decade, a new technique of SBVS called consensus virtual screening (CVS) has been used in some studies to increase the accuracy of SBVS and to reduce the false positives obtained in these experiments. An indispensable condition to be able to utilize SBVS is the availability of a 3D structure of the target protein. Some virtual databases, such as the Protein Data Bank, have been created to store the 3D structures of molecules. However, sometimes it is not possible to experimentally obtain the 3D structure. In this situation, the homology modeling methodology allows the prediction of the 3D structure of a protein from its amino acid sequence. This review presents an overview of the challenges involved in the use of CADD to perform SBVS, the areas where CADD tools support SBVS, a comparison between the most commonly used tools, and the techniques currently used in an attempt to reduce the time and cost in the drug development process. Finally, the final considerations demonstrate the importance of using SBVS in the drug development process.

Molecular Architect: A User-Friendly Workflow for Virtual Screening

Computer-assisted drug design (CADD) methods have greatly contributed to the development of new drugs. Among CADD methodologies, virtual screening (VS) can enrich the compound collection with molecules that have the desired physicochemical and pharmacophoric characteristics that are needed to become drugs. Many free tools are available for this purpose, but they are difficult to use and do not have a graphical user interface. Furthermore, several free tools must be used to carry out the entire VS process, requiring the user to process the results of one software program so that they can be used in another program, adding a potential source of human error. Moreover, some software programs require knowledge of advanced computational skills, such as programming languages. This context has motivated us to develop Molecular Architect (MolAr). MolAr is a workflow with a simple and intuitive interface that acts in an integrated and automated form to perform the entire VS process, from protein preparation (homology modeling and protonation state) to virtual screening. MolAr carries out VS through AutoDock Vina, DOCK 6, or a consensus of the two. Two case studies were conducted to demonstrate the performance of MolAr. In the first study, the feasibility of using MolAr for DNA-ligand systems was assessed. Both AutoDock Vina and DOCK 6 showed good results in performing VS in DNA-ligand systems. However, the use of consensus virtual screening was able to enrich the results. According to the area under the ROC curve and the enrichment factors, consensus VS was better able to predict the positions of the active ligands. The second case study was performed on 8 targets from the DUD-E database and 10 active ligands for each target. The results demonstrated that using the final ligand conformation provided by AutoDock Vina as an input for DOCK 6 improved the DOCK 6 ROC curves by up to 42% in VS. These case studies demonstrated that MolAr is capable conducting the VS process and is an easy-to-use and effective tool. MolAr is available for download free of charge at http: //www.drugdiscovery.com.br/software/.

Deep Transferable Compound Representation across Domains and Tasks for Low Data Drug Discovery

The main problem of small molecule-based drug discovery is to find a candidate molecule with increased pharmacological activity, proper ADME, and low toxicity. Recently, machine learning has driven a significant contribution to drug discovery. However, many machine learning methods, such as deep learning-based approaches, require a large amount of training data to form accurate predictions for unseen data. In lead optimization step, the amount of available biological data on small molecule compounds is low, which makes it a challenging problem to apply machine learning methods. The main goal of this study is to design a new approach to handle these situations. To this end, source assay (auxiliary assay) knowledge is utilized to learn a better model to predict the property of new compounds in the target assay. Up to now, the current approaches did not consider that source and target assays are adapted to different target groups with different compounds distribution. In this paper, we propose a new architecture by utilizing graph convolutional network and adversarial domain adaptation network to tackle this issue. To evaluate the proposed approach, we applied it to Tox21, ToxCast, SIDER, HIV, and BACE collections. The results showed the effectiveness of the proposed approach in transferring the related knowledge from source to target data set.

Hispaglabridin B, a constituent of liquorice identified by a bioinformatics and machine learning approach, relieves protein-energy wasting by inhibiting forkhead box O1

BACKGROUND AND PURPOSE

Liquorice is the root of Glycyrrhiza glabra, which is a popular food in Europe and China that has previously shown benefits for skeletal fatigue and nutrient metabolism. However, the mechanism and active ingredients remain largely unclear. The aim of this study was to investigate the active ingredients of liquorice for muscle wasting and elucidate the underlying mechanisms.

EXPERIMENTAL APPROACH

RNA-Seq and bioinformatics analysis were applied to predict the main target of liquorice. A machine learning model and a docking tool were used to predict active ingredients. Isotope labelling experiments, immunostaining, Western blots, qRT-PCR, ChIP-PCR and luciferase reporters were utilized to test the pharmacological effects in vitro and in vivo. The reverse effects were verified through recombination-based overexpression.

KEY RESULTS

The liposoluble constituents of liquorice improved muscle wasting by inhibiting protein catabolism and fibre atrophy. We further identified FoxO1 as the target of liposoluble constituents of liquorice. In addition, hispaglabridin B (HB) was predicted as an inhibitor of FoxO1. Further studies determined that HB improved muscle wasting by inhibiting catabolism in vivo and in vitro. HB also markedly suppressed the transcriptional activity of FoxO1, with decreased expression of the muscle-specific E3 ubiquitin ligases MuRF1 and Atrogin-1.

CONCLUSIONS AND IMPLICATIONS

HB can serve as a novel natural food extract for preventing muscle wasting in chronic kidney disease and possibly other catabolic conditions.