Efficient molecular encoders for virtual screening

Molecular representations encoding molecular structure information play critical roles in molecular virtual screening (VS). In order to improve VS performance, an abundance of molecular encoders have been developed and tested by various VS challenges. Combinational strategies were also used to improve the performance. Deep learning (DL)-based molecular encoders have attracted much attention for their automatic information extraction ability. In this review, we present an overview of two-dimensional-, three-dimensional-, and DL-based molecular encoders, summarize recent progress of VS using DL technologies, and propose a general framework of DL molecular encoder-based VS. Perspectives on the future directions of molecular representations and applications in the prediction of active compounds are also provided.     

Detailed analysis of scoring functions for virtual screening

We present a comprehensive study of the performance of fast scoring functions for library docking using the program FlexX as the docking engine. Four scoring functions, among them two recently developed knowledge-based potentials, are evaluated on seven target proteins whose binding sites represent a wide range of size, form, and polarity. The results of these calculations give valuable insight into strengths and weaknesses of current scoring functions. Furthermore, it is shown that a well-chosen combination of two of the tested scoring functions leads to a new, robust scoring scheme with superior performance in virtual screening.     

Enhancing preclinical drug discovery with artificial intelligence

Artificial intelligence (AI) is becoming an integral part of drug discovery. It has the potential to deliver across the drug discovery and development value chain, starting from target identification and reaching through clinical development. In this review, we provide an overview of current AI technologies and a glimpse of how AI is reimagining preclinical drug discovery by highlighting examples where AI has made a real impact. Considering the excitement and hyperbole surrounding AI in drug discovery, we aim to present a realistic view by discussing both opportunities and challenges in adopting AI in drug discovery.     

A bioinformatics search for selective histamine h4 receptor antagonists through structure-based virtual screening strategies

The prevalence of allergic disease is increasing dramatically in the developed world. Studies of allergic diseases have clearly demonstrated that histamine plays an important role in the pathogenesis of the early-phase allergic response. Histamine effects are mediated by H1, H2, H3, and H4 receptors. The presence of the histamine H4 receptors on leukocytes and mast cells suggests that the new histamine receptor H4 plays an important role in the modulation of the immune system. Thus, histamine H4 receptor is an attractive target for anti-allergic therapy. In our present study, we have generated a histamine H4 receptor model using I-TASSER based on human B2-adrenergic G-protein-coupled receptor. Structurally similar compounds of the three known antagonists JNJ777120, thioperamide, and Vuf6002 were retrieved from PubChem, and database was prepared. Virtual screening of those databases was performed, and six compounds with high docking score were identified. Also the binding mode revealed that all the six compounds had interaction with Asp94 of the receptor. Our results serve as a starting point in the development of novel lead compounds in anti-allergic therapy.     

Combining ligand-based and structure-based drug design in the virtual screening arena

The aim of virtual high-throughput screening is the identification of biologically relevant molecules among either tangible or virtual (large) collections of compounds. Likewise, high-throughput screening (HTS) and high-throughput virtual screening (HTVS) methods are becoming very important within the drug discovery process. HTVS methods can be categorised as either ‘ligand-based’ or ‘structure-based’ depending on if a direct knowledge of the three-dimensional target structure is required. A summary of the most promising computational approaches is reviewed. Advantages and shortcomings of the methodology are also discussed.     

Identification of new classes of ricin toxin inhibitors by virtual screening

We used two virtual screening programs, ICM and GOLD, to dock nearly 50,000 compounds into each of two conformations of the target protein ricin A chain (RTA). A limited control set suggests that candidates scored highly by two programs may have a higher probability of being ligands than those in a list from a single program. Based on the virtual screens, we purchased 306 compounds that were subjected to a kinetic assay. Six compounds were found to give modest, but significant, inhibition of RTA. They also tended to inhibit Shiga toxin A chain, with roughly the same IC₅₀. The compounds generally represent novel chemical platforms that do not resemble RTA substrates, as currently known inhibitors do. These six were also tested in a cell-based assay for their ability to protect cells from intact ricin. Two compounds were effective in this regard, showing modest to strong ricin inhibition, but also showing some cytotoxicity. RTA, with its large, polar active site is a difficult drug design target which is expected to bind small molecules only weakly. The ability of the method to find these novel platforms is encouraging and suggests virtual screening can contribute to the search for ricin and Shiga toxin inhibitors.     

Modifications of the scoring function in FlexX for virtual screening applications

A modification of the hydrogen bond score in the docking program FlexX is presented. Hydrogen bonds formed in inaccessible regions of protein cavities thereby gain larger weight than others formed at the protein surface. The modified scoring function is tested with thrombin as a target. Secondly, a recently published knowledge-based scoring function is comparedto the FlexX scoring function in several database ranking experiments. This revised version was published online in June 2006 with corrections to the Cover Date.     

新冠病毒SARS-CoV-2主蛋白酶3CLpro抑制剂的多步虚拟筛选

目的 通过构建多步虚拟筛选模型和方法,寻找有效筛选新冠状病毒SARS-CoV-2主蛋白酶3CLpro抑制剂的方法,为抗击COVID-19提供去潜在的候选药物。方法 利用Discovery Studio Client 2016和AutoDock Vina 1.5.6软件对ZINC 数据库共5811个化合物进行基于结构的虚拟筛选,然后基于配体的药效团模型,选取最佳模型后进行基于配体的虚拟筛选,最后基于构效关系筛选出与抗病毒有关的药物。结果 经过3步筛选最终得到11个具有高亲和力、高拟合值和适当药理作用的化合物。结论 通过3种不同方法的多步筛选可以快速寻找出具有高活性的候选化合物,表明构建的虚拟筛选的方法和模型有效,为今后抗COVID-19药物研究提供了潜在的研究工具。     

New approaches to QSAR: Neural networks and machine learning

Summary Neural networks and machine learning are two methods that are increasingly being used to model QSARs. They make few statistical assumptions and are nonlinear and nonparametric. We describe back-propagation from the field of neural networks, and GOLEM from machine learning, and illustrate their learning mechanisms using a simple expository problem. Back-propagation and GOLEM are then compared with multiple linear regression (using the parameters and their squares) on two real drug design problems: the inhibition ofEscherichia coli dihydrofolate reductase (DHFR) by pyrimidines and the inhibition of rat/mouse tumour DHFR by triazines.     

Predicting drug metabolism and pharmacokinetics features of in-house compounds by a hybrid machine-learning model

We constructed machine learning-based pharmacokinetic prediction models with very high performance. The models were trained on 26138 and 16613 compounds involved in metabolic stability and cytochrome P450 inhibition, respectively. Because the compound features largely differed between the publicly available and in-house compounds, the models learned on the public data could not predict the in-house compounds, suggesting that outside of a certain applicability domain (AD), the prediction results are unreliable and can mislead the design of novel compounds. To exclude the uncertain prediction results, we constructed another machine learning model that determines whether the newly designed compound is inside or outside the AD. The AD was evaluated multi-dimensionally with some explanatory variables: The structural similarities and the probability obtained from the pharmacokinetic prediction model. The accuracy of predicting metabolic stability was 79.9% on the test set, increasing significantly to 93.6% after excluding the low-reliability compounds. The model properly classified the reliability of the compounds. After learning on the in-house compounds, the reliability model classified almost all (90%) of the public compounds as low reliability, indicating that the AD was properly determined by the model.     

Identification of potential inhibitors of Zika virus NS5 RNA-dependent RNA polymerase through virtual screening and molecular dynamic simulations

Zika virus (ZIKV) is one of the mosquito borne flavivirus with several outbreaks in past few years in tropical and subtropical regions. The non-structural proteins of flaviviruses are suitable active targets for inhibitory drugs due to their role in pathogenicity. In ZIKV, the non-structural protein 5 (NS5) RNA-Dependent RNA polymerase replicates its genome. Here we have performed virtual screening to identify suitable ligands that can potentially halt the ZIKV NS5 RNA dependent RNA polymerase (RdRp). During this process, we searched and screened a library of ligands against ZIKV NS5 RdRp. The selected ligands with significant binding energy and ligand-receptor interactions were further processed. Among the selected docked conformations, top five was further optimized at atomic level using molecular dynamic simulations followed by binding free energy calculations. The interactions of ligands with the target structure of ZIKV RdRp revealed that they form strong bonds within the active sites of the receptor molecule. The efficacy of these drugs against ZIKV can be further analyzed through in-vitro and in-vivo studies.     

Current methods and challenges for deep learning in drug discovery

Driven by rapid advances in computer hardware and publicly available datasets over the past decade, deep learning has achieved tremendous success in the transformation of many computational disciplines. These novel technologies have had considerable impact on computer-aided drug design as well, throughout all stages of the development pipeline. A flexible toolbox of neural architectures has been developed that are well-suited to represent the sequential, topological, or geometrical concepts of chemistry and biology; and that are able to either discriminate existing molecules or to generate new ones from scratch. For some biochemical prediction tasks, the state of the art has been advanced; however, for complex and practically relevant projects, the outcomes are less clear-cut. Current deep learning methods rely on massive amounts of labeled examples, but drug discovery data is comparatively limited in quantity and quality. These problems need to be resolved and existing sources used more effectively to demonstrate that deep learning can revolutionize the field in general.     

Exploiting PubChem for virtual screening

Importance of the field: PubChem is a public molecular information repository, a scientific showcase of the National Institutes of Health Roadmap Initiative. The PubChem database holds > 27 million records of unique chemical structures of compounds (compound ID) derived from nearly 70 million substance depositions (substance ID), and contains > 449,000 bioassay records with thousands of in vitro biochemical and cell-based screening bioassays established targeting > 7000 proteins and genes linking to > 1.8 million of substances.

Areas covered in this review: This review builds on recent PubChem-related computational chemistry research reported by other authors while providing readers with an overview of the PubChem database, focusing on its increasing role in cheminformatics, virtual screening and toxicity prediction modeling.

What the reader will gain: These publicly available data sets in PubChem provide great opportunities for scientists to perform cheminformatics and virtual screening research for computer-aided drug design. However, the high volume and complexity of the data sets, in particular the bioassay-associated false positives/negatives and highly imbalanced data sets in PubChem, also create major challenges. Several approaches regarding the modeling of PubChem data sets and development of virtual screening models for bioactivity and toxicity predictions are also reviewed.

Take home message: Novel data-mining cheminformatics tools and virtual screening algorithms are being developed and used to retrieve, annotate and analyze the large-scale and highly complex PubChem biological screening data for drug design.     

基于结构的肽脱甲酰基酶抑制剂的虚拟筛选及活性研究

目的通过计算机虚拟筛选寻找肽脱甲酰基酶(Peptide deformylase,PDF)的小分子抑制剂。方法从PDB(Protein Data Bank)数据库中获得与肠球菌(Enteroccous faecium)PDF序列相似性最高的肺炎链球菌PDF的X-射线衍射晶体结构,应用Sybyl7.3软件中的Sur?ex-Dock对我室微生物天然产物数据库(Microbial Natural Products Database,MNPD)中约15000个样品进行虚拟筛选;采用高通量荧光检测法和微稀释法分别测定虚拟筛选获得的部分高得分化合物对PDF酶的抑制作用和抗菌活性。结果应用Surflex-Dock虚拟筛选获得得分达5.17及以上的命中化合物651个,经综合分析,挑选出其中5个化合物进行进一步的体外生物学活性研究,结果表明,1个化合物(spergualin)有一定的PDF抑制活性和抗菌活性,其它4个化合物仅有弱活性或无活性。结论发现化合物spergualin具有一定的PDF抑制活性和抗耐药表皮葡萄球菌活性。证实了计算机虚拟筛选与实物筛选相结合可明显提高PDF酶抑制剂的筛选效率。     

药效团模型法寻找M1受体激动剂

寻找新的M1受体激动剂先导化合物。在M1受体三维结构未知的情况下，利用距离比较法（DISCO）将10个结构特征具有代表性的M1受体激动剂的分子构象进行叠合，建立了可能的药效团模型，初步验证了该模型的可靠性。利用该模型对ACD-SC数据库进行虚拟筛选，购买了22个与药效团叠合较好、与已知M1受体激动剂结构类型不同的化合物，并对其进行活性测定。结果发现了一个具有M1受体激动活性的化合物，其EC50为4．90μmol／L，最大响应倍数为10．0，值得进行更深入研究。     

Abiraterone Acetate Attenuates SARS-CoV-2 Replication by Interfering with the Structural Nucleocapsid Protein

The drug repurposing strategy has been applied to the development of emergency COVID-19 therapeutic medicines. Current drug repurposing approaches have been directed against RNA polymerases and viral proteases. Recently, we found that the inhibition of the interaction between the SARS-CoV-2 structural nucleocapsid (N) and spike (S) proteins decreased viral replication. In this study, drug repurposing candidates were screened by in silico molecular docking simulation with the SARS-CoV-2 structural N protein. In the ChEMBL database, 1994 FDA-approved drugs were selected for the in silico virtual screening against the N terminal domain (NTD) of the SARS-CoV-2 N protein. The tyrosine 109 residue in the NTD of the N protein was used as the center of the ligand binding grid for the docking simulation. In plaque forming assays performed with SARS-CoV-2 infected Vero E6 cells, atovaquone, abiraterone acetate, and digoxin exhibited a tendency to reduce the size of the viral plagues without affecting the plaque numbers. Abiraterone acetate significantly decreased the accumulation of viral particles in the cell culture supernatants in a concentration-dependent manner. In addition, abiraterone acetate significantly decreased the production of N protein and S protein in the SARS-CoV-2-infected Vero E6 cells. In conclusion, abiraterone acetate has therapeutic potential to inhibit the viral replication of SARS-CoV-2.     

Using machine learning for improving knowledge on antibacterial effect of bioactive glass

The aim of this work was to find relationships between critical bioactive glass characteristics and their antibacterial behaviour using an artificial intelligence tool. A large dataset including ingredients and process variables of the bioactive glasses production, bacterial characteristics and microbiological experimental conditions was generated from literature and analyzed by neurofuzzy logic technology. Our findings allow an explanation on the variability in antibacterial behaviour found by different authors and to obtain general conclusions about critical parameters of bioactive glasses to be considered in order to achieve activity against some of the most common skin and implant surgery pathogens.     

Identification of putative estrogen receptor-mediated endocrine disrupting chemicals using QSAR- and structure-based virtual screening approaches

Identification of endocrine disrupting chemicals is one of the important goals of environmental chemical hazard screening. We report on the development of validated in silico predictors of chemicals likely to cause estrogen receptor (ER)-mediated endocrine disruption to facilitate their prioritization for future screening. A database of relative binding affinity of a large number of ERα and/or ERβ ligands was assembled (546 for ERα and 137 for ERβ). Both single-task learning (STL) and multi-task learning (MTL) continuous quantitative structure–activity relationship (QSAR) models were developed for predicting ligand binding affinity to ERα or ERβ. High predictive accuracy was achieved for ERα binding affinity (MTL R² = 0.71, STL R² = 0.73). For ERβ binding affinity, MTL models were significantly more predictive (R² = 0.53, p < 0.05) than STL models. In addition, docking studies were performed on a set of ER agonists/antagonists (67 agonists and 39 antagonists for ERα, 48 agonists and 32 antagonists for ERβ, supplemented by putative decoys/non-binders) using the following ER structures (in complexes with respective ligands) retrieved from the Protein Data Bank: ERα agonist (PDB ID: 1L2I), ERα antagonist (PDB ID: 3DT3), ERβ agonist (PDB ID: 2NV7), and ERβ antagonist (PDB ID: 1L2J). We found that all four ER conformations discriminated their corresponding ligands from presumed non-binders. Finally, both QSAR models and ER structures were employed in parallel to virtually screen several large libraries of environmental chemicals to derive a ligand- and structure-based prioritized list of putative estrogenic compounds to be used for in vitro and in vivo experimental validation.