Computational Approaches in Biomolecular Modeling and Characterization: A Comprehensive Overview

Computational Medicinal Chemistry: Part II

Current computational methods for predicting protein interactions of natural products

Computational Medicinal Chemistry

Computer-aided drug design and discovery methods have been essential in developing small molecules with therapeutic properties over the last decades. Application of computational resources includes drug target identification, hit discovery, and lead optimization. Accordingly, with tremendous research efforts and the availability of financial support from government agencies across the world, and multinational drug companies, the overall research level in this area will continue to advance. The methodology used in this review paper entailed a thorough examination of research studies on relevant literature on drug design and development using computational resources. Extensive searches using Scopus, International Pharmaceutical s (OvidSp, WHO Global Health Library, Cochrane, Google Scholar, Web of Science, Science Direct, ProQuest dissertation & theses, Worldwide Political Science s (CSA), and PubMed was carried out. A standardized template was used to ensure that the selected papers met the inclusion criteria, and relevant to the review. Ultimately, there are robust technologies developed to enhance the drug discovery process. Therefore, this review provides insights into computational resources in Silico and ab initio methods and algorithms, not restricted to drug metabolism predictions for drug design, and the practical applications of artificial intelligence (AI) in drug discovery. Computational tools and methods for drug design and development such as molecular dynamics (MD), molecular docking, quantum mechanics (QM), hybrid quantum mechanics/molecular mechanics (QM/MM), and Density functional theory (DFT) have been reviewed. Accordingly, the emerging technique of synergistically employing these techniques influences the fundamental challenges of conventional medicines for complex diseases. Herein, we discuss ligand-based and structure-based drug discoveries, force field models in MD simulations, docking algorithms, subtractive and additive QM/MM coupling. Nonetheless, as computer-aided drug (CADD) approaches continue to evolve with significant improvements, the focus areas will be on docking and virtual screening, scoring functions, optimization of hits, and assessment of adsorption, distribution, metabolism, excretion, and toxicity (ADMET) properties. With the current success, the present computational resources will aid in the future discovery of novel compounds with high therapeutic performance. The ongoing oncology research efforts will also significantly contribute to UN sustainable development goals – good health and well-being, sustainable innovation and industrialization.

It all clicks together: In silico drug discovery becoming mainstream

This editorial introduces a new special focus issue dedicated to computational chemistry and computer-aided drug discovery to be published in the third quarter of 2016, for which the author will serve as a guest editor. To position this special issue some background information is provided and a few important aspects are highlighted. Medicinal chemistry publications frequently contain computational studies. Many of these studies are descriptive in nature. For example, a computational model is added to an experimental structure–activity relationship (SAR) investigation. This typically leads to the formulation of hypotheses as to how a compound might interact with its target or why active compounds differ in their potency. Another popular exercise is virtual screening aiming to identify novel active compounds using structureand/or ligand-based computational approaches. Less frequent are computational investigations reporting the prospective design of new compounds for synthesis and biological evaluation. In addition, compound property analyses are carried out, for example, to rationalize or predict drug-likeness or ADME characteristics. Furthermore, new computational concepts or methods are published to aid in drug discovery efforts at different levels. Computational studies reported in the medicinal chemistry literature are often characterized by a high degree of scientific heterogeneity. For example, modeling of compound binding modes is a popular exercise, often carried out to rationalize SARs. However, such predictions are frequently over-interpreted and putative interactions are discussed at the atomic level of detail as if they were experimental observations. On the other hand, there are also careful studies that evaluate putative binding modes taking the accuracy limitations of modeling into account and formulate experimentally testable hypotheses. However, the situation becomes particularly delicate when ‘multihypothetical’ strategies are pursued. This is illustrated by considering other examples from structure-based modeling. It is not uncommon that homology models are built for targets of interest, active compounds are docked into modeled binding sites, binding (free) energies are calculated for these ‘double-hypothetical’ ligand–target complexes, and correlated with experimentally determined affinities. What can one conclude from such exercises if close correlation between computed binding energies and experimental data is ultimately reported? Do such findings ‘validate’ the computational approach? Or might they represent the ‘luck of the draw’? Scientific views might well differ in such cases. Regardless, these types of investigations more or less follow a ‘look what computational methods can do!’ theme and often give incorrect impressions, especially to non-experts. This is far from being helpful for the further development of the field at Pushing the boundaries of computational approaches: special focus issue on computational chemistry and computer-aided drug discovery

Chapter 2 - Computational Drug Design Methods—Current and Future Perspectives

The ultimate goal of a QSAR analysis is prediction, which depends on the elaboration of the most appropriate set of molecular descriptors. As such, molecular description is the nucleus of QSAR and in the absence of exhaustive molecular description, rational drug design may be greatly impeded. As previously discussed, computational methods such as quantum mechanics and molecular mechanics provide molecular description at a fundamental level which then enhances the descriptive capability and predictive power of a QSAR analysis. In recognition of these capabilities, semi-empirical molecular orbital methods and molecular mechanics now have been incorporated into or interphased with QSAR programs. Such integrated packages are being successfully used in computer-aided molecular modeling. Computer-aided molecular modeling can provide the three-dimensional structure of a molecule, its chemical and physical characteristics, comparisons of structures of different molecules, and visualization of complexes formed between them. From the foregoing, predictions may be made about how related new molecules may function. Thus, the combination of quantum and/or molecular mechanics and QSAR provides a formidable weapon in the chemist's armamentarium. The molecular modeling approaches are certainly more practical to use than physicochemical methods. They also provide electronic and thermodynamic data that are not available from x-ray crystallographic data. Of course, these techniques are not confined to radiopharmaceutical development and they also could aid in the development of contrast agents for radiography or magnetic resonance imaging. We believe that as computational resources and capabilities increase over the next decade, computer-aided drug design will become a standard procedure in all drug development laboratories.

Background Masticatory Myofascial Pain Dysfunction Syndrome (MMPDS) is a musculoligamentous disorder that shares similarities with temporomandibular joint pain and odontogenic pain. It manifests as dull or aching pain in masticatory muscles, influenced by jaw movement. Computer-aided drug design (CADD) encompasses various theoretical and computational approaches used in modern drug discovery. Molecular docking is a prominent method in CADD that facilitates the understanding of drug-bimolecular interactions for rational drug design, mechanistic studies & the formation of stable complexes with increased specificity and potential efficacy. The docking technique provides valuable insights into binding energy, free energy, and complex stability predictions. Aim The aim of this study was to use the docking technique for myosin inhibitors. Materials and methods Four inhibitors of myosin were chosen from the literature. These compound structures were retrieved from the Zinc15 database. Myosin protein was chosen as the target and was optimized using the RCSB Protein Data Bank. After pharmacophore modeling, 20 novel compounds were found and the SwissDock was used to dock them with the target protein. We compared the binding energies of the newly discovered compounds to those of the previously published molecules with the target. Results The results indicated that among the 20 molecules ZINC035924607 and ZINC5110352 exhibited the highest binding energy and displayed superior properties compared to the other molecules. Conclusion The study concluded that ZINC035924607 and ZINC5110352 exhibited greater binding affinity than the reported inhibitors of myosin. Therefore, these two molecules can be used as a potential and promisingleadfor the treatment of MMPDSand could be employed in targeted drug therapy.

There has been impressive growth in the use of radiopharmaceuticals for therapy, selective toxic payload delivery, and noninvasive diagnostic imaging of disease. The increasing timeframes and costs involved in the discovery and development of new radiopharmaceuticals have driven the development of more efficient strategies for this process. Computer-Aided Drug Design (CADD) methods and Machine Learning (ML) have become more effective over the last two decades for drug and materials discovery and optimization. They are now fast, flexible, and sufficiently accurate to accelerate the discovery of new molecules and materials. Radiopharmaceuticals have also started to benefit from rapid developments in computational methods. Here, we review the types of computational molecular design techniques that have been used for radiopharmaceuticals design. We also provide a thorough examination of success stories in the design of radiopharmaceuticals, and the strengths and weaknesses of the computational methods. We begin by providing a brief overview of therapeutic and diagnostic radiopharmaceuticals and the steps involved in radiopharmaceuticals design and development. We then review the computational design methods used in radiopharmaceutical studies, including molecular mechanics, quantum mechanics, molecular dynamics, molecular docking, pharmacophore modelling, and datadriven ML. Finally, the difficulties and opportunities presented by radiopharmaceutical modelling are highlighted. The review emphasizes the potential of computational design methods to accelerate the production of these very useful clinical radiopharmaceutical agents and aims to raise awareness among radiopharmaceutical researchers about computational modelling and simulation methods that can be of benefit to this field.

Electrochemistry in Ionic Liquid (IL) electrolytes is an area of research with significant opportunity for innovation and of high relevance for environmental and energy applications. While computational chemistry methods exist and have continuously evolved over recent decades and contributed significantly to providing complementary information and insight for experimental physical science, there is relatively limited knowledge about applying them for electrochemical problems. This is due to the complication of the heterogeneous electrochemical interface and system. It is more challenging to apply computational methods to understand electrode/IL interface reactions and processes since ILs and/or IL containing electrolytes, owing to their unique nature of ions interacting with other ions, analytes, solvent molecules, and electrodes, add additional complexity to the current methods. This dissertation addresses important aspects of this challenge by taking three topics of electrochemistry with IL-based electrolytes and systematically leverages computational tools to advance the fundamental understanding of these IL based electrochemical systems. This approach leads to advancements of theoretical considerations regarding IL electrochemistry and to progress in the practical implementation of respective calculations. The first research subject is to understand the detail mechanisms of the dissociative reduction of trichloroethylene (TCE), which is important for TCE remediation. Computational approaches are proposed and used for the study of reaction pathways of TCE in an IL/acetonitrile mixed electrolyte. Systematic considerations of the thermodynamics, kinetics, and electrolyte structure related issues are included in the computational study. The second research subject is understanding the CO2 physical chemistry such as conductivity, mass density, and structure of the IL and their impact on CO2 solubility and adsorption that are valuable for the development of real time and continuous CO2 sensor applications. Computational methods are utilized to understand the mechanism that leads to selective impedance variation upon CO2 and methane exposure. The third research topic is dedicated to further computational methods for the study of CO2 and N2O reduction in IL electrolytes. Bulk, and surface structures of electrolyte, as well as the reaction pathways of CO2 and N2O reductions in these electrochemical systems are investigated, and novel methods to study gaseous molecules' redox chemistry in the IL electrolytes are developed, based on computational chemistry. With each of the model problems based on IL and/or IL containing electrolytes, calculation methods were proposed, implemented, and refined. The results of this work contribute to the advancement of computational chemistry to the complex field of IL and electrode interface electrochemistry, which offers solutions to critical environmental and energy related challenges.

As part of a large medicinal chemistry program, we wish to develop novel selective estrogen receptor modulators (SERMs) as potential breast cancer treatments using a combination of experimental and computational approaches. However, one of the remaining difficulties nowadays is to fully integrate computational (i.e., virtual, theoretical) and medicinal (i.e., experimental, intuitive) chemistry to take advantage of the full potential of both. For this purpose, we have developed a Web-based platform, Forecaster, and a number of programs (e.g., Prepare, React, Select) with the aim of combining computational chemistry and medicinal chemistry expertise to facilitate drug discovery and development and more specifically to integrate synthesis into computer-aided drug design. In our quest for potent SERMs, this platform was used to build virtual combinatorial libraries, filter and extract a highly diverse library from the NCI database, and dock them to the estrogen receptor (ER), with all of these steps being fully automated by computational chemists for use by medicinal chemists. As a result, virtual screening of a diverse library seeded with active compounds followed by a search for analogs yielded an enrichment factor of 129, with 98% of the seeded active compounds recovered, while the screening of a designed virtual combinatorial library including known actives yielded an area under the receiver operating characteristic (AU-ROC) of 0.78. The lead optimization proved less successful, further demonstrating the challenge to simulate structure activity relationship studies.

Traditional drug design requires a great amount of research time and developmental expense. Booming computational approaches, including computational biology, computer-aided drug design, and artificial intelligence, have the potential to expedite the efficiency of drug discovery by minimizing the time and financial cost. In recent years, computational approaches are being widely used to improve the efficacy and effectiveness of drug discovery and pipeline, leading to the approval of plenty of new drugs for marketing. The present review emphasizes on the applications of these indispensable computational approaches in aiding target identification, lead discovery, and lead optimization. Some challenges of using these approaches for drug design are also discussed. Moreover, we propose a methodology for integrating various computational techniques into new drug discovery and design.

Computer-aided drug design (CADD), especially artificial intelligence-driven drug design (AIDD), is increasingly used in drug discovery. In this paper, a novel and efficient workflow for hit identification was developed within the ID4Inno drug discovery platform, featuring innovative artificial intelligence, high-accuracy computational chemistry, and high-performance cloud computing. The workflow was validated by discovering a few potent hit compounds (best IC50 is ∼0.80 μM) against PI5P4K-β, a novel anti-cancer target. Furthermore, by applying the tools implemented in ID4Inno, we managed to optimize these hit compounds and finally obtained five hit series with different scaffolds, all of which showed high activity against PI5P4K-β. These results demonstrate the effectiveness of ID4inno in driving hit identification based on artificial intelligence, computational chemistry, and cloud computing.

Computational approaches have promised to organize collections of functional genomics data into testable predictions of gene and protein involvement in biological processes and pathways. However, few such predictions have been experimentally validated on a large scale, leaving many bioinformatic methods unproven and underutilized in the biology community. Further, it remains unclear what biological concerns should be taken into account when using computational methods to drive real-world experimental efforts. To investigate these concerns and to establish the utility of computational predictions of gene function, we experimentally tested hundreds of predictions generated from an ensemble of three complementary methods for the process of mitochondrial organization and biogenesis in Saccharomyces cerevisiae. The biological data with respect to the mitochondria are presented in a companion manuscript published in PLoS Genetics (doi:10.1371/journal.pgen.1000407). Here we analyze and explore the results of this study that are broadly applicable for computationalists applying gene function prediction techniques, including a new experimental comparison with 48 genes representing the genomic background. Our study leads to several conclusions that are important to consider when driving laboratory investigations using computational prediction approaches. While most genes in yeast are already known to participate in at least one biological process, we confirm that genes with known functions can still be strong candidates for annotation of additional gene functions. We find that different analysis techniques and different underlying data can both greatly affect the types of functional predictions produced by computational methods. This diversity allows an ensemble of techniques to substantially broaden the biological scope and breadth of predictions. We also find that performing prediction and validation steps iteratively allows us to more completely characterize a biological area of interest. While this study focused on a specific functional area in yeast, many of these observations may be useful in the contexts of other processes and organisms.