Quantum Deep Learning Functional Similarities on Remdesivir, Drug Synergies to Treat COVID-19 in Practice.

Νovel SARS coronavirus 2 (SARS-CoV-2) of the family Coronaviridae starting in China and spreading around the world is an enveloped, positive-sense, single-stranded RNA of the genus betacoronavirus encoding the SARS-COV-2 (2019-NCOV, Coronavirus Disease 2019. Remdesivir drug, or GS-5734 lead compound, rst described in 2016 as a potential anti-viral agent for Ebola diseade and has also being researched as a potential therapeutic agent against the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the coronavirus that causes coronavirus disease 2019 (COVID-19). Computer-aided drug design (CADD), Structure and Ligand based Drug Repositioning strategies based on parallel docking methodologies have been widely used for both modern drug development and drug repurposing to nd effective treatments against this disease. Quantum mechanics, molecular mechanics, molecular dynamics (MD), and combinations have shown superior performance to other drug design approaches providing an unprecedented opportunity in the rational drug development elds and for the developing of innovative drug repositioning methods. We tested 18 phytochemical small molecule libraries and predicted their synergies in COVID-19 (2019- NCOV), to devise therapeutic strategies, repurpose existing ones in order to counteract highly pathogenic SARS-CoV-2 infection and the BRD4- conserved residue associated COVID-19 pathology. We anticipate that our quantum deep learing similarity approaches can be used for the development of anticoronaviral drug combinations in large scale HTS screenings, and to maximize the safety and ecacy of the Remdesivir, Colchicine and Ursolic acid drugs already known to induce synergy with potential therapeutic value or drug repositioning to COVID-19 patients. .pdbqt and .mol2 format les (26,27-42,46-48) present in the e-Drug3D dataset. (28,30-48) The ensemble of 3D molecule conformations (31,32-45,48) of the larger drugs from e-Drug3D were provided on a separated drug library dataset.


Drugs under clinical trials (COVID-19 repurposing dataset)
Our approach was focused on identifying a cluster of (1-48) similar chemotypes followed by parallel docking grid generation using the advanced to the next MM-PBSA-WSAS, KNIME-HTS-HVS lter, and chemical structure preparation wizard as provided by the BiogenetoligandorolTM cluster of algorithms that had the potential to target the (2-48) FURIN-ADAMTS1-ROR-GAMMA-SARS-COV-2 conserved domains and t the geometric constraints without any restraints or constraints when lling in the open valence of the SARS-COV-2-ACE2-RORγ-BRD4-FURIN (1-30,41-48) binding pocket residues. These were then (6-48) converted into substructure searches of one copy of SARS-CoV-2 main protease (7-45) which were used to mine commercially available  compounds and covalently bonded SARS-COV-2 inhibitors using the (9-32) eMolecules database. The dataset of the selected hit drugs followed the force eld parameters as applied to the partial atomic charges of the selectred ligands which were derived using the RESP and collected from published articles (3,5,7,13,48) and approved drugs listed on the (2-39) DrugBank database in the "Clinical Trial Summary by Drug" section to t the HF/6-31G* electrostatic potentials generated using the Gaussian 16 software package. (4,(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18) We intended to generate these from two different branches to model the viral protein using the isotropic position scaling algorithm. (7,48) One branch of the selected hit elements was refered to as the "Remdesivir Literature Substructures" branch, (8, which was based on the Remdesivir, Colchicine and Ursolic acid substructures using the Antechamber module as extracted (15,46) from published bromodomain inhibitors.

Discussion
In this research report, we found that the Cluster of the Recombovir-(Drug Combination) which were identi ed during screening of a compound diversity set performed by the BiogenetoligandorolTM cluster of algorithms on the intersection track (Lys711 and Arg355/SARS-CoV2 PLpro and Lys711 and Arg355. The chemical strucutures of the Remdesivir, Colchicine and Ursolic acid) targeted into the Lys711 and Arg355 residues and inside the residues of the Phe19, Trp23, and Leu26, which are located in an alphahelical region of the SARS-CoV2 PLpro N terminus that binds to the N-terminal Lys711 and Arg355 hydrophobic pocket (17). The druggable scaffold of this drug combination of the Remdesivir, Colchicine and Ursolic acid small molecules target into the binding domaisn of these three critical SARS-CoV2 PLpro residues; the combination of the hit compounds therefore competes with endogenous SARS-CoV2 PLpro for binding to Lys711 and Arg355. We created a new track to display the contacts of this drug combination with each of those: Lys711 and Arg355, and SARS-CoV2 PLpro. Interestingly, this drug combination consisted of the Remdesivir, Colchicine and Ursolic acid chemical structures targets the Lys711 and Arg355 homo-dimerization site and intersects within the Lys711 and Arg355-Recombovir-(Drug Combination) binding sites, suggesting that they may also interfere within the binding pockets of the Lys711 and Arg355 homodimerizations.The key residues of the seven hot spot residues that contribute to S spike glycoprotein trimerization can be potentially In-Silico dow-regulated by the docking combination of the therapeutic agents of the Remdesivir, Colchicine and Ursolic acid drugs in order to signi cantly block the quaternary structure assembly of the SARS-CoV-2 replication protein. Furthermore, the combination of the drugs of the Remdesivir, Colchicine and Ursolic acid targets the BRD4, JQ-1, ISG15, IFN-β, IL-1β, I-BET 151 and OTX-015 exhibited viral inhibition with a short hairpin RNA (shRNA; shBRD4) which is involved in the recognition of molecular patterns and mediate severe in ammatory responses, may also interact within the binding domains of the S protein via 10 active residues located in the S1 subunit. I also suggested that the Recomborovir-(Drug Combination) might interact with this ganglioside-binding domain within the S protein (61). Drug repurposing or the chemical optimization of existing drugs represent an effective drug discovery approach and Drug Combination therapeutic approach which has the potential to reduce the time and costs associated to the de novo drug discovery and development of this anti-COVID-19 clinical trial process (62-72,80-94). This In-silico project demonstrated that the combination of the drugs of the Remdesivir, Colchicine and Ursolic acid can potentially inhibit the SARS-CoV-2 replication (63), (64). Additionally, the BiogenetoligandorolTM algorithm cannot determine binding free energies or binding orientations of small molecules. For that aim, other docking tools or molecular dynamics studies should be applied, as explained above. Therefore, the BiogenetoligandorolTM approach that takes a multiple-sequence-conserved alignment (coMSA) in .pdbqt format le as an input is mainly aimed at binding residues recognition in cases on QMMM homology modeling techniques where the binding partner is a small chemical compound or small peptide. Its druggability to predict docking tness scoring effectiveness allowing us to generate this drug repurposing screening approach by combining its cluster free energy ranking output with other chemistry informatics and repositioning In-Silico tools. Therefore, it can be used as an AI-strategy in complex inverse docking and quantum simulation pipelines. We envision the BiogenetoligandorolTM quantum thinking procedure as the rst step in a ligand parallel and inverse docking and free energy simulation  (-400.794, 329.678, -337.184, -907.342, -52.667, -894.194, -194 Therefore, solutions provided by the BiogenetoligandorolTM cluster of AI-Algorithms in this project indicated to us that the Colchicine, Remdesivir and Ursolic acid drugs are considered to be <<co-administered>> (Figures3e, 4a, 4b, 4c, 4d, 4e, 4f, 4g) which is something more than important and have to be considered as a rst approximation that may require subsequent parallel re nement and docking analysis using more accurate free energy ranking models. In conclusion, BiogenetoligandorolTM -LigandorolTM is very e cientand is not just proposed as an alternative drug repurposing and computational method, but rather as a combined complementary deep learning similarity and quantum mechanics predictive tool to be used in tandem with other In-Silico drug retargeting and computational platforms which could led us to the rational design of novel drug combinations of small molecules and more effective repositioning experimental methods.

Declarations
Availability of data and materials The author con rms that the data supporting the ndings of this study are available within the article [and/or] its supplementary materials.

Competing interests
No potential competing interest was reported by the author.

Funding
The author received no nancial support for the research, authorship, and/or publication of this article.

Authors' contributions
Author's diverse contributions to the published work are accurate and agreed.
Author has contributed in the below multiple roles: Ø Conceptualization Ideas, Formulation or evolution of overarching research goals and aims Ø Methodology, Development and design of methodology; creation of models Ø Software, Programming, software development; designing computer programs; implementation of the computer code and supporting algorithms; testing of existing code components Ø Validation, Veri cation, whether as a part of the activity or separate, of the overall replication/ reproducibility of results/experiments and other research outputs Ø Formal analysis Application of statistical, mathematical, computational, or other formal techniques to analyze or synthesize study data Ø Investigation, Conducting a research and investigation process, speci cally performing the experiments, or data/evidence collection Ø Resources, Provision of study materials, reagents, materials, patients, laboratory samples, animals, instrumentation, computing resources, or other analysis tools Ø Data Curation, Management activities to annotate (produce metadata), scrub data and maintain research data (including software code, where it is necessary for interpreting the data itself) for initial use and later reuse Ø Writing -Original Draft, Preparation, creation and presentation of the published work, speci cally writing the initial draft (including substantive translation) Ø Writing -Review & Editing Preparation, creation and/or presentation of the published work by those from the original research group, speci cally critical review, commentary or revision -including pre-or postpublication stages Ø Visualization, Preparation, creation and presentation of the published work, speci cally visualization/ data presentation Ø Supervision, Oversight and leadership responsibility for the research activity planning and execution, including mentorship external to the core team Ø Project administration, Management and coordination responsibility for the research activity planning and execution. Acknowledgments I would like to deeply express my special thanks of gratitude to my teacher (George Grigoriadis Pharmacist) as well as our CEO and principal (Nikolaos Grigoriadis Phd Pharmacist) who gave me the golden opportunity to do this wonderful project on the Quantum Deep Learning Chemistry topic, which also helped me in doing a lot of Original Drug Repurposing and Drug Combination Research and I came to know about so many new things I am really thankful to them.

Signi cance Statement
Drug repurposing/repositioning/rescue proposed a computational method to identify potential drug indications by integrating various applications of an existing drug to a new disease indication. In this paper we ltered out residues with relatively small surface accessible areas, and/or with incompatible charge and hydrophobic properties to the ligands of the Remdesivir, Colchicine and Ursolic acid small molecules which could improve the prediction binding free energies or binding orientations of different drug combinations of the Remdesivir, Colchicine and Ursolic acid to treat COVID-19. Finally, an comprehensive web platform by applying AI deep learning models was designed based on our BiogenetoligandorolTM protocol for drug repurposing to signi cantly reduce user time for data gathering and multi-step analysis without human intervention.In conclusion, BiogenetoligandorolTM -LigandorolTM is not proposed as an alternative drug repurposing method, but rather as a complementary deep learning quantum mechanics tool to be used in tandem with other drug retargeting computational and small molecule repositioning experimental methods.