Virtual Drug Design Research Articles

Trends of Artificial Intelligence (AI) Use in Drug Targets, Discovery and Development: Current Status and Future Perspectives.

The applications of artificial intelligence (AI) in pharmaceutical sectors have advanced drug discovery and development methods. AI has been applied in virtual drug design, molecule synthesis, advanced research, various screening methods, and decision-making processes. In the fourth industrial revolution, when medical discoveries are happening swiftly, AI technology is essential to reduce the costs, effort, and time in the pharmaceutical industry. Further, it will aid "genome-based medicine" and "drug discovery." AI may prepare proactive databases according to diseases, disorders, and appropriate usage of drugs which will facilitate the required data for the process of drug development. The application of AI has improved clinical trials on patient selection in a population, stratification, and sample assessment such as biomarkers, effectiveness measures, dosage selection, and trial length. Various studies suggest AI could be perform better compared to conventional techniques in drug discovery. The present review focused on the positive impact of AI in drug discovery and development processes in the pharmaceutical industry and beneficial usage in health sectors as well.

APMG: 3D Molecule Generation Driven by Atomic Chemical Properties.

Recently, mask-fill-based 3D Molecular Generation (MG) methods have become very popular in virtual drug design. However, the existing MG methods ignore the chemical properties of atoms and contain inappropriate atomic position training data, which limits their generation capability. To mitigate the above issues, this paper presents a novel mask-fill-based 3D molecule generation model driven by atomic chemical properties (APMG). Specifically, we construct a new attention-MPNN-based encoder and introduce the electronic information into atom representations to enrich chemical properties. Also, a multi-functional classifier is designed to predict the electronic information of each generated atom, guiding the type prediction of elements and bonds. By design, the proposed method uses the chemical properties of atoms and their correlations for high-quality molecule generation. Second, to optimize the atomic position training data, we propose a novel atomic training position generation approach using the Chi-Square distribution. We evaluate our APMG method on the CrossDocked dataset and visualize the docking states of the pockets and generated molecules. The obtained results demonstrate the superiority and merits of APMG over the state-of-the-art approaches. The dataset and codes will be available on the project homepage: https://github.com/JU-HuaY/APMG.

The Discovery of a Novel Antimetastatic Bcl3 Inhibitor.

The development of antimetastatic drugs is an urgent healthcare priority for patients with cancer, because metastasis is thought to account for around 90% of cancer deaths. Current antimetastatic treatment options are limited and often associated with poor long-term survival and systemic toxicities. Bcl3, a facilitator protein of the NF-κB family, is associated with poor prognosis in a range of tumor types. Bcl3 has been directly implicated in the metastasis of tumor cells, yet is well tolerated when constitutively deleted in murine models, making it a promising therapeutic target. Here, we describe the identification and characterization of the first small-molecule Bcl3 inhibitor, by using a virtual drug design and screening approach against a computational model of the Bcl3-NF-kB1(p50) protein-protein interaction. From selected virtual screening hits, one compound (JS6) showed potent intracellular Bcl3-inhibitory activity. JS6 treatment led to reductions in Bcl3-NF-kB1 binding, tumor colony formation, and cancer cell migration in vitro; and tumor stasis and antimetastatic activity in vivo, while being devoid of overt systemic toxicity. These results represent a successful application of in silico screening in the identification of protein-protein inhibitors for novel intracellular targets, and confirm Bcl3 as a potential antimetastatic target.

Eltrombopag is a potential target for drug intervention in SARS-CoV-2 spike protein

The COVID-19 pandemic, caused by the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), is a current global threat for which there is an urgent need to search for an effective therapy. The transmembrane spike (S) glycoprotein of SARS-CoV-2 directly binds to the host angiotensin-converting enzyme 2 (ACE2) and mediates viral entrance, which is therefore considered as a promising drug target. Considering that new drug development is a time-consuming process, drug repositioning may facilitate rapid drug discovery dealing with sudden infectious diseases. Here, we compared the differences between the virtual structural proteins of SARS-CoV-2 and SARS-CoV, and selected a pocket mainly localizing in the fusion cores of S2 domain for drug screening. A virtual drug design algorithm screened the Food and Drug Administration-approved drug library of 1234 compounds, and 13 top scored compounds were obtained through manual screening. Through in vitro molecular interaction experiments, eltrombopag was further verified to possess a high binding affinity to S protein plus human ACE2 and could potentially affect the stability of the ACE2-S protein complex. Hence, it is worth further exploring eltrombopag as a potential drug for the treatment of SARS-CoV-2 infection.

Identifying new topoisomerase II poison scaffolds by combining publicly available toxicity data and 2D/3D-based virtual screening

Molecular descriptor (2D) and three dimensional (3D) shape based similarity methods are widely used in ligand based virtual drug design. In the present study pairwise structure comparisons among a set of 4858 DTP compounds tested in the NCI60 tumor cell line anticancer drug screen were computed using chemical hashed fingerprints and 3D molecule shapes to calculate 2D and 3D similarities, respectively. Additionally, pairwise biological activity similarities were calculated by correlating the 60 element vectors of pGI50 values corresponding to the cytotoxicity of the compounds across the NCI60 panel. Subsequently, we compared the power of 2D and 3D structural similarity metrics to predict the toxicity pattern of compounds. We found that while the positive predictive value and sensitivity of 3D and molecular descriptor based approaches to predict biological activity are similar, a subset of molecule pairs yielded contradictory results. By simultaneously requiring similarity of biological activities and 3D shapes, and dissimilarity of molecular descriptor based comparisons, we identify pairs of scaffold hopping candidates displaying characteristic core structural changes such as heteroatom/heterocycle change and ring closure. Attempts to discover scaffold hopping candidates of mitoxantrone recovered known Topoisomerase II (Top2) inhibitors, and also predicted new, previously unknown chemotypes possessing in vitro Top2 inhibitory activity.

Computer-Aided Drug Design in Epigenetics.

Epigenetic dysfunction has been widely implicated in several diseases especially cancers thus highlights the therapeutic potential for chemical interventions in this field. With rapid development of computational methodologies and high-performance computational resources, computer-aided drug design has emerged as a promising strategy to speed up epigenetic drug discovery. Herein, we make a brief overview of major computational methods reported in the literature including druggability prediction, virtual screening, homology modeling, scaffold hopping, pharmacophore modeling, molecular dynamics simulations, quantum chemistry calculation, and 3D quantitative structure activity relationship that have been successfully applied in the design and discovery of epi-drugs and epi-probes. Finally, we discuss about major limitations of current virtual drug design strategies in epigenetics drug discovery and future directions in this field.

Grid-based Continual Analysis of Molecular Interior for Drug Discovery, QSAR and QSPR.

In 1979, R.D.Cramer and M.Milne made a first realization of 3D comparison of molecules by aligning them in space and by mapping their molecular fields to a 3D grid. Further, this approach was developed as the DYLOMMS (Dynamic Lattice- Oriented Molecular Modelling System) approach. In 1984, H.Wold and S.Wold proposed the use of partial least squares (PLS) analysis, instead of principal component analysis, to correlate the field values with biological activities. Then, in 1988, the method which was called CoMFA (Comparative Molecular Field Analysis) was introduced and the appropriate software became commercially available. Since 1988, a lot of 3D QSAR methods, algorithms and their modifications are introduced for solving of virtual drug discovery problems (e.g., CoMSIA, CoMMA, HINT, HASL, GOLPE, GRID, PARM, Raptor, BiS, CiS, ConGO,). All the methods can be divided into two groups (classes):1. Methods studying the exterior of molecules; 2) Methods studying the interior of molecules. A series of grid-based computational technologies for Continual Molecular Interior analysis (CoMIn) are invented in the current paper. The grid-based analysis is fulfilled by means of a lattice construction analogously to many other grid-based methods. The further continual elucidation of molecular structure is performed in various ways. (i) In terms of intermolecular interactions potentials. This can be represented as a superposition of Coulomb, Van der Waals interactions and hydrogen bonds. All the potentials are well known continual functions and their values can be determined in all lattice points for a molecule. (ii) In the terms of quantum functions such as electron density distribution, Laplacian and Hamiltonian of electron density distribution, potential energy distribution, the highest occupied and the lowest unoccupied molecular orbitals distribution and their superposition. To reduce time of calculations using quantum methods based on the first principles, an original quantum free-orbital approach AlteQ is proposed. All the functions can be calculated using a quantum approach at a sufficient level of theory and their values can be determined in all lattice points for a molecule. Then, the molecules of a dataset can be superimposed in the lattice for the maximal coincidence (or minimal deviations) of the potentials (i) or the quantum functions (ii). The methods and criteria of the superimposition are discussed. After that a functional relationship between biological activity or property and characteristics of potentials (i) or functions (ii) is created. The methods of the quantitative relationship construction are discussed. New approaches for rational virtual drug design based on the intermolecular potentials and quantum functions are invented. All the invented methods are realized at www.chemosophia.com web page. Therefore, a set of 3D QSAR approaches for continual molecular interior study giving a lot of opportunities for virtual drug discovery, virtual screening and ligand-based drug design are invented. The continual elucidation of molecular structure is performed in the terms of intermolecular interactions potentials and in the terms of quantum functions such as electron density distribution, Laplacian and Hamiltonian of electron density distribution, potential energy distribution, the highest occupied and the lowest unoccupied molecular orbitals distribution and their superposition. To reduce time of calculations using quantum methods based on the first principles, an original quantum free-orbital approach AlteQ is proposed. The methods of the quantitative relationship construction are discussed. New approaches for rational virtual drug design based on the intermolecular potentials and quantum functions are invented. All the invented methods are realized at www.chemosophia.com web page.

Prospects of Applying Enhanced Semi-Empirical QM Methods for 2101 Virtual Drug Design

The last five years have seen a renaissance of semiempirical quantum mechanical (SQM) methods in the field of virtual drug design, largely due to the increased accuracy of so-called enhanced SQM approaches. These methods make use of additional terms for treating dispersion (D) and hydrogen bond (H) interactions with an accuracy comparable to dispersion-corrected density functional theory (DFT-D). DFT-D in turn was shown to provide an accuracy comparable to the most sophisticated QM approaches when it comes to non-covalent intermolecular forces, which usually dominate the protein/ligand interactions that are central to virtual drug design. Enhanced SQM methods thus offer a very promising way to improve upon the current state of the art in the field of virtual drug design.

Virtual drug design: Skp1–Skp2 inhibition targets cancer stem cells

The dysregulation of pathways regulating cellular function is a frequent hallmark of cancer and the development of specific pathway inhibitors that alter tumor growth and progression are the focus of multiple recent studies. E3 ubiquitin ligases are a large group of diverse protein enzymes that specifically target proteins for clearance, and their importance to normal cellular function is illustrated in the many diseases associated with their loss of function or inappropriate targeting. S-phase kinase-associated protein 2 (Skp2) is an F box protein that plays critical roles in cell-cycle progression, senescence, metabolism, and acts as an Skp1–Cullin-1–F box (SCF) ubiquitin ligase substrate recognition factor. Overexpression of Skp2 is associated with poor prognosis and metastasis in many cancers and is a well validated drug target. In a recent report, Chan et al. have identified an Skp2 inhibitor that selectively impairs Skp2 E3 ligase activity using an integrated virtual high-throughput drug screening and experimental validation approach. This Skp2 inhibitor restricts cancer stemness and potentiates sensitivity to chemotherapeutic agents in multiple animal tumor models. These findings identify a new novel small molecule that targets the Skp2 and reduces tumor growth by attenuating aerobic glycolysis and inducing cellular senescence.

Comparison of Molecular Mechanics, Semi-Empirical Quantum Mechanical, and Density Functional Theory Methods for Scoring Protein–Ligand Interactions

Correctly ranking protein-ligand interactions with respect to overall free energy of binding is a grand challenge for virtual drug design. Here we compare the performance of various quantum chemical approaches for tackling this so-called "scoring" problem. Relying on systematically generated benchmark sets of large protein/ligand model complexes based on the PDBbind database, we show that the performance depends first of all on the general level of theory. Comparing classical molecular mechanics (MM), semiempirical quantum mechanical (SQM), and density functional theory (DFT) based methods, we find that enhanced SQM approaches perform very similar to DFT methods and substantially different from MM potentials.

Abstract 2076: Inhibiting glycolysis and HIF1α: a novel approach to cancer therapy.

Abstract Cancer cells satisfy their increased need for energy under conditions of normal oxygen tension by undergoing a high rate of glycolysis followed by lactic acid fermentation, a process known as the Warburg Effect, rather than oxidative phosphorylation as observed in normal cells. Unlike proliferating cancer cells, the internal environment of most solid tumors is hypoxic due to an inadequate or disorganized blood supply. This results in tumors generating ATP through anaerobic respiration, a process many times less efficient than oxidative phosphorylation. The importance of anaerobic glycolysis for cancer cell energy generation makes it an attractive target for inhibiting cancer growth. However, agents currently in use for regulating energy production in tumors are relatively ineffective analogs of glucose or intermediates of the glycolytic pathway. We performed a high throughput siRNA screen using an siRNA library targeting all known open reading frames of the human genome (Dharmacon Inc.) to identify genes that when inhibited would block hypoxia inducible factor-1 alpha (HIF-1α) transcription factor luciferase reporter activity. A significant number of genes selected and validated were members of the glycolysis pathway. It is known that many glycolysis enzymes are induced by HIF-1α providing a feed forward loop between glycolysis and HIF-1α. We have addressed two questions: (1) what is the biological mechanism by which these two signaling systems regulate one another?; and (2) can specific inhibitors of glycolysis be developed with the added effect of inhibiting HIF-1α? Our mechanistic studies have shown that of the 40 enzymes of glycolysis involving 11 enzymatic steps and various enzyme isoforms, siRNA knockdown of Aldolase A (ALDOA) produces the maximum inhibition of HIF-1α activity. Further work indicated that ALDOA siRNA inhibition of HIF-1α is mediated through AMPK, a sensor of low ATP levels in the cell, since siRNA knockdown of AMPK results in rescue of HIF-1α activity. The effect of ALDOA inhibition is not mediated through a reduction in HIF-1α protein suggesting that, contrary to the canonical AMPK pathway, mTOR and its regulation of translation is not involved. Instead, p300 is phosphorylated by AMPK thus inhibiting its function as a co-activator of HIF-1α. A structure based drug design approach was employed for lead identification and optimization using the crystal structure of ALDOA (4ALD) and GOLD virtual screening platform. Three compounds have been identified as having effects on the modulation of ALDOA activity in a protein based biochemical assay as well as inhibitory effects on glycolysis itself using the XF96 Analyzer from Seahorse Bioscience. Here, we present an overview of ALDOA target identification, its mechanism as a regulator of HIF-1α activity and virtual drug design methodologies used in the search for novel pharmacological probes which act as dual inhibitors of energy production and the HIF-1α survival pathway. Citation Format: Geoffrey V. Grandjean, John Kingston, John Kenneth Morrow, Shuxing Zhang, Garth Powis. Inhibiting glycolysis and HIF1α: a novel approach to cancer therapy. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 2076. doi:10.1158/1538-7445.AM2013-2076

Ligand-binding site prediction using ligand-interacting and binding site-enriched protein triangles

Knowledge about the site at which a ligand binds provides an important clue for predicting the function of a protein and is also often a prerequisite for performing docking computations in virtual drug design and screening. We have previously shown that certain ligand-interacting triangles of protein atoms, called protein triangles, tend to occur more frequently at ligand-binding sites than at other parts of the protein. In this work, we describe a new ligand-binding site prediction method that was developed based on binding site-enriched protein triangles. The new method was tested on 2 benchmark datasets and on 19 targets from two recent community-based studies of such predictions, and excellent results were obtained. Where comparisons were made, the success rates for the new method for the first predicted site were significantly better than methods that are not a meta-predictor. Further examination showed that, for most of the unsuccessful predictions, the pocket of the ligand-binding site was identified, but not the site itself, whereas for some others, the failure was not due to the method itself but due to the use of an incorrect biological unit in the structure examined, although using correct biological units would not necessarily improve the prediction success rates. These results suggest that the new method is a valuable new addition to a suite of existing structure-based bioinformatics tools for studies of molecular recognition and related functions of proteins in post-genomics research. The executable binaries and a web server for our method are available from http://sourceforge.net/projects/msdock/ and http://lise.ibms.sinica.edu.tw, respectively, free for academic users.

Translational Research

Translational Research

Relating the shape of protein binding sites to binding affinity profiles: is there an association?

BackgroundVarious pattern-based methods exist that use in vitro or in silico affinity profiles for classification and functional examination of proteins. Nevertheless, the connection between the protein affinity profiles and the structural characteristics of the binding sites is still unclear. Our aim was to investigate the association between virtual drug screening results (calculated binding free energy values) and the geometry of protein binding sites. Molecular Affinity Fingerprints (MAFs) were determined for 154 proteins based on their molecular docking energy results for 1,255 FDA-approved drugs. Protein binding site geometries were characterized by 420 PocketPicker descriptors. The basic underlying component structure of MAFs and binding site geometries, respectively, were examined by principal component analysis; association between principal components extracted from these two sets of variables was then investigated by canonical correlation and redundancy analyses.ResultsPCA analysis of the MAF variables provided 30 factors which explained 71.4% of the total variance of the energy values while 13 factors were obtained from the PocketPicker descriptors which cumulatively explained 94.1% of the total variance. Canonical correlation analysis resulted in 3 statistically significant canonical factor pairs with correlation values of 0.87, 0.84 and 0.77, respectively. Redundancy analysis indicated that PocketPicker descriptor factors explain 6.9% of the variance of the MAF factor set while MAF factors explain 15.9% of the total variance of PocketPicker descriptor factors. Based on the salient structures of the factor pairs, we identified a clear-cut association between the shape and bulkiness of the drug molecules and the protein binding site descriptors.ConclusionsThis is the first study to investigate complex multivariate associations between affinity profiles and the geometric properties of protein binding sites. We found that, except for few specific cases, the shapes of the binding pockets have relatively low weights in the determination of the affinity profiles of proteins. Since the MAF profile is closely related to the target specificity of ligand binding sites we can conclude that the shape of the binding site is not a pivotal factor in selecting drug targets. Nonetheless, based on strong specific associations between certain MAF profiles and specific geometric descriptors we identified, the shapes of the binding sites do have a crucial role in virtual drug design for certain drug categories, including morphine derivatives, benzodiazepines, barbiturates and antihistamines.

Model of a specific human histamine H3 receptor (hH3R) binding pocket suitable for virtual drug design

Model of a specific human histamine H3 receptor (hH3R) binding pocket suitable for virtual drug design