A Novel Artificial Intelligence and Machine learning-based scoring system for evaluating re-purposing potential of Valproic Acid drug in COVID-19.
Akanksha Aggarwal1, Prashant Singh2, Robin Sinha2, Preeti P2, Trapti Sharma2, Swarsat Kaushik Nath2, Kamal Rawal#1
1Delhi Institute of Pharmaceutical Sciences and Research, Delhi Pharmaceutical Sciences and Research University, New Delhi, 110017, India
2Amity Institute of Biotechnology, Amity University Uttar Pradesh, India.
#1Corresponding Author Email ID: kamal.rawal@gmail.com
Centre for Computational Biology and Bioinformatics, AIB Amity University, Noida
Background
Despite widespread vaccinations and the introduction of many repurposed medications, the rise of COVID-19 reinfection due to the SARS-CoV Omicron (B.1.1.529) variant has presented a significant challenge to health authorities across the world. There is a critical need for novel healthcare medications. Valproic acid (VPA) has been reported to be a beneficial drug due to its excellent property to hinder enveloped viral multiplication. Artificial Intelligence (AI) and Machine Learning (ML) have been used extensively to predict the repurposing potential of a drug by examining its past couple of years’ activity. The strategies for drug repositioning that will play a substantial role in this approach can be widely categorized into AI approaches, network-based models, and structure-based approaches.
Methodology
We have implemented a multimodal pipeline that operates on computational and machine learning strategies, namely molecular docking, molecular data, chemical information, clinical data, and gene expression to analyze the drugs for their sensitivity against covid-19. Here, we implemented the COV-DrugX pipeline to comprehend the physicochemical properties of VPA interactions with viral protein (targets), identify a place in the gene expression profile, additionally its potential role in human network databases.
Result and Conclusion
Here, we have analyzed the ML pipeline developed by our research group (Rawal et al., 2021) to identify the COVID-associated drug repurposing properties through AI and learning prediction module and use the COV-DrugX pipeline, that predicts the repurposing properties between different existing COVID-19 drugs. Nsp13 was found to have the highest binding affinity (-5.6 kcal/mol) with VPA. By using different modules, we have found that VPA displays physicochemical, biological, and other characteristics similar to existent COVID-19 drugs.
Keyword
Molecular Docking; Valproic Acid; VPA; Bioinformatics; drug repurposing; artificial intelligence; COVID-19; SAR-CoV-2
CoV-DrugX Pipeline | Home (kamalrawal. in) and Tools (kamalrawal. in)
At present, Coronavirus Disease 2019 (COVID-19) pandemic, caused by the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2), has led to major oppression of worldwide healthcare infrastructure [Lai et al., 2020]. Widespread efforts are being put in by the scientists day in and day out to identify appropriate therapies against the viral disease. Although vaccine development and therapeutic antibodies are being industriously investigated, an increasing number of such promising approaches are the need of the hour, because they may lead us to continued efficacy in treating COVID-19 [Lu et al., 2020]. However, its inception has also resulted in a considerable number of clinical trials that assessed combinations of drugs made up of repurposed therapies ( references for drugs trials of VPA - NCT04513314 ).
VPA is a well-known histone deacetylases (HDAC2) inhibitor, that has received widespread acceptance for its usage in the treatment of epilepsy and other neurological illnesses during the last 50 years [Sébastien et al., 2010]. Recent data suggest that VPA may be useful in the treatment of some malignancies, due to its involvement in modifying epigenetic alterations by inhibiting histone deacetylases, which directly or indirectly influence the expression of genes related to the cell cycle, differentiation, and death [JE et al., 2020].VPA's direct impact on immune system cells has just lately been investigated. VPA affects the suppression of various activation mechanisms in different immune cells that contribute to an anti-inflammatory response is discussed in this study [Kong et al., 2020]. VPA displays its influences on the immune cells by demonstrating the expression of the cell cycle and death genes via epigenetic alterations [MedlinePlus Drug Information, n.d.]. VPA stimulates RNA interference, triggers histone methyltransferases, as well as restrains transcription factor activation along with blocking histone deacetylases [Li et al., 2020]. However, the effectiveness of VPA during the infectious phase is determined by the pathogen's biological characteristics and the related immune response; this is because VPA can support infection control or advancement [Soria-Castro et al.,2019]. VPA, because of its diverse effects, is a promising option for autoimmune diseases and hypersensitivity that is worth further investigation [Soria-Castro et al.,2019].
VPA has been shown to inhibit enveloped viral multiplication [Vazquez-Calvo A et al., 2011] and decreased levels of inflammatory cytokines in human monocyte-derived macrophages infected with Dengue virus e.g. tumor necrosis factor-alpha (TNF- alpha) and Interleukin-6 (IL-6) [Delgado F.G et al., 2018]. IL-6 levels were discovered to be a major biomarker of illness severity and death (illustrated in Figure 1) [Wang F et al., 2020]. VPA has been proven to alter macrophage responses to lipopolysaccharide (LPS) by decreasing IL-12p70 and TNF- production while raising IL-10 [Haskó et al., 2000]. Valproic acid can reduce angiotensin-converting enzyme 2(ACE2) and transmembrane serine protease 2 (TMPRSS2) production [MedlinePlus Drug Information, n.d.], all of which are essential for SARS-CoV-2 viral entry, ( see Figure 2) as well as supervise the immune cells and cytokine response to infection, minimizing end-organ damage [Pitt et al., 2021].
Previously, our research team has developed a number of machine learning and bioinformatics systems. Text mining and network biology-based systems are two examples [Jagannadham et al., 2016], likewise vaccine discovery system [Rawal et al. 2021], next-generation sequencing analysis system for cancer and other genomes [Preeti et al.,2021, Rawal et al 2011, Mandal et al 2006]. Moreover, they recently developed the first comprehensive pipeline and machine learning-based resource for analyzing mobile genomic elements and epigenetic mutations in different disorders [Rawal et al., 2011]. By merging large-scale experimental data with computational and comparative approaches, researchers were able to construct an unexplored methodology for determining the function of microRNAs in heart development and disorders. We discovered 353 previously unknown and 703 new miRNAs involved in heart maturation. Genes associated with the cell cycle, extracellular remodeling, apoptosis, metabolism, signaling pathways, transcriptional regulators, and chromatin remodeling seemed to be enriched in the target mRNAs [Rustagi et al., 2015], [Gupta, R., Soni, et al., 2010],[Abbasi, B. A. et al., 2020]
Many scholars have formed multiple new frameworks/portals for "obesity treatment", additional for diseases such as type 2 diabetes and hypertension amongst sedentary individuals by including important sections of the community over the use of social networks, data science systems, machine learning, sensors, and mobile apps. The technology is being used as a social service venture by the general population [K. Rawal, P. Gaur et al., 2014], for instance, the creation of a web-based weight-loss program[K. Rawal, S. Agarwal, et al., 2014]. Moreover, the advancement of the first molecular network on human obesity over screening >25 million PubMed records, for clinical studies, drug side effects, gene expression databases, and other information resources progressed the way for more computational bioinformatics research. Like the development of a machine learning model and a new text mining system for semi-automated screening of literature records with higher F scores [Jagannadham, J., Jaiswal, 2016].
We have designed a drug repurposing tool that examines the suitability of a drug (e.g. VPA) with existing FDA-approved drugs in COVID-19 treatment. Its algorithm is based on 14 different modules namely - CoV-DrugX, CoV-DrugX SE, CoV-DrugX Phenotype, CoV-DrugX Target, CoV-DrugX Condition, CoV-DrugX Symptoms, CoV-DrugX Gene Expressions, CoV-DrugX Pathway circuit, CoV-DrugX AI Ranking, CoV-DrugX Knowledge Graph, Drug-DL 11, Drug-DL 200, Drug Dock, Drug Dock Batch [Rawal et al., 2021].
Therefore, we run the pipeline to inspect the repurposing property for VPA. The docking module will give us the data on binding and interaction between COVID-19 proteins and the drug that we had given from the docking results we can determine whether the drug is suitable for repurposing or not.
For the examination, choose a particular drug, VPA, to check its re-purposing property. Further, we checked for its drug bank ID (we used DrugBank) and curated the isomeric smiles of it from the section chemical identifiers from the website. we saved the isomeric smiles in the .txt formats in the system. Next, we submitted its SMILES [CCCC(CCC)C(O)=O] as a query in the AI module (drugX pipeline), and one can choose some filters according to needs (see Figure 4 and Figure 5 ).
The results can be saved in CSV format and can be investigated accordingly (see Figure 6).
The second module (Table 2) suggests that the database includes data regarding drug COVID-19 repurposing 200 properties. For a more thorough assessment, all null values have been deleted from the acquired data source. Table 3 depicts the positive results that the given drug has shown interaction with various COVID-19 targets and has a number of the common targets too. Table 4 implies that the given drug has shown abnormal human gene expression during COVID-19. In response to the gene interaction, some of the drugs exhibited up-regulation (AKR1B7, HDAC7), though some of them showed down-regulation (HDAC2, CYP3A4), and in a couple of cases, a gene can be expressed either way depending on the factors unspecified (ALDH5A1, AKR1B1).
Another approach we used in our pipeline is docking-based predictions and we had built 3 docking-based modules where we have a human target-based docking module, viral proteins target-based module, and protein kinases associated with human proteins as targets for the query drugs.
When docked against 23 viral proteins, using drug_dock_viral, results are interpreted in the form of Binding Affinities (which shows the strength of binding interaction). The average binding affinity came out to be -4.30 Kcal/mol. Amongst all, the 5 highest interacting viral proteins were Nsp13, Nsp14, Npro, Nsp15, S_trimer with -5.6, -5.6, -5.1, -5.1, -5.0, respectively as their binding energies in kcal/mol. While, on the contrary, the lowest binding energy was observed in the case of Nsp1 and Orf10 with -2.6 and -3.2 Kcal/mol. See Table 5 and Figure 7 to understand data collected and a histogram showing the distribution of binding affinities against viral proteins. Table 6 and Figure 8 represents data for the docking results of VPA when docked against Human proteins (ACE2 and TMPRSS2)and associated protein kinases (GAK, JAK12, and AAK1) in form of binding affinities in kcal/mol.
We can conclude from Table 7 and Figure 9 that the drug has shown response with the SARS-CoV-2 pathway circuit. Table 8 reports that our database has data regarding having euclidean distance between drug and SARS-CoV-2 disease. Although, in a couple of cases, VPA showed COVID-19 side effects but doesn’t have any COVID-19 phenotype. However, it doesn’t cure any COVID-19 conditions.
We compiled the score of all 10 modules in the binary form of 0 and 1, except the docking modules where we found the results in this range of 0 and 1, taking into consideration the collective score for the docked proteins in the docking module (see Table 1). After analyzing VPA with the Cov-DrugX system, we've found 10 modules (out of 13 available) that had some information relevant to the drug, and out of which, 7 module reports a binary score of 1 (70%) and 3 modules reports a score of 0 (the modules Drug 11 properties, COVID-19 drug_Condition, and COVID-19 drug_phenotype) because of lacking any evident information about the drug in the source databases associated with these modules as there has been no gene expression associated to the drug reported yet. Hence, we can recognize the drug to be evaluated as a COVID-19 repurposable drug.
We can conclude from our analysis that VPA, a histone deacetylases (HDAC2) inhibitor has high capabilities to fight against the COVID-19 virus, and since filter studies had already proved that VPA and Nsp13 showed excellent binding affinity (-5.6 kcal/mol); should be able to treat COVID-19, is a promising alternative for COVID-19 repurposing drug treatment, as an effective, cheap, and easy to use drug which can hit all the targets. Clinical trials can help to find VPA’s role in governing the control of autoimmune diseases and hypersensitivity reactions that can be further explored, to discover more about the mechanism of action and characteristics features.
Lai CC, Shih TP, Ko WC, Tang HJ, Hsueh PR. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and coronavirus disease-2019 (COVID-19): The epidemic and the challenges. International journal of antimicrobial agents. 2020 Mar 1;55(3):105924.
Wang H, Zheng X, Jin J, Zheng L, Guan T, Huo Y, Xie S, Wu Y, Chen W. LncRNA MALAT1 silencing protects against cerebral ischemia-reperfusion injury through miR-145 to regulate AQP4. Journal of biomedical science. 2020 Dec;27(1):1-2.
Kong FC, Ma CL, Zhong MK. Epigenetic effects mediated by antiepileptic drugs and their potential application. Current neuropharmacology. 2020 Feb 1;18(2):153-66.
Audia JE, Campbell RM. Histone modifications and cancer. Cold Spring Harbor perspectives in biology. 2016 Apr 1;8(4):a019521.
Chateauvieux S, Morceau F, Dicato M, Diederich M. Molecular and therapeutic potential and toxicity of valproic acid. Journal of Biomedicine and Biotechnology. 2010 Jul 29;2010.
Li G, Tian Y, Zhu WG. The roles of histone deacetylases and their inhibitors in cancer therapy. Frontiers in Cell and Developmental Biology. 2020 Sep 29;8:1004.
Valproic Acid: MedlinePlus Drug Information. Medlineplus.gov. Retrieved from National Library of Medicine
Soria-Castro R, Schcolnik-Cabrera A, Rodríguez-López G, Campillo-Navarro M, Puebla-Osorio N, Estrada-Parra S, Estrada-García I, Chacón-Salinas R, Chávez-Blanco AD. Exploring the drug repurposing versatility of valproic acid as a multifunctional regulator of innate and adaptive immune cells. Journal of immunology research. 2019 Mar 14;2019.
Vázquez-Calvo Á, Saiz JC, Sobrino F, Martín-Acebes MA. Inhibition of enveloped virus infection of cultured cells by valproic acid. Journal of virology. 2011 Feb 1;85(3):1267-74.
Delgado FG, Cárdenas P, Castellanos JE. Valproic acid downregulates cytokine expression in human macrophages infected with dengue virus. Diseases. 2018 Sep;6(3):59.
Wang F, Hou H, Luo Y, Tang G, Wu S, Huang M, Liu W, Zhu Y, Lin Q, Mao L, Fang M. The laboratory tests and host immunity of COVID-19 patients with different severity of illness. JCI insight. 2020 May 21;5(10).
Hasko G. Kuhel DG, Salzman AL, Szabo C. ATP suppression of interleukin-12 and tumor decreased necrosis factor-α release from macrophages. Br J Pharmacol. 2000;129:909-14.
Drugs, H. (2021). Valproic Acid: MedlinePlus Drug Information. Medlineplus.gov. Retrieved 13 December 2021, from https://medlineplus.gov/druginfo/meds/a682412.html.
Pitt B, Sutton NR, Wang Z, Goonewardena SN, Holinstat M. Potential repurposing of the HDAC inhibitor valproic acid for patients with COVID-19. European journal of pharmacology. 2021 Mar 2:173988.
CoV-DrugX Pipeline | Home. Drugx.kamalrawal.in. (2021). Retrieved 13 December 2021, from http://drugx.kamalrawal.in/drugx/
Rawal, K. and Ramaswamy, R., "Genome-wide analysis of mobile genetic elements insertion sites. Nucl. Acids Res.,vol. 39, no. 16, pp. 6864-6878, Sep. 2011.
Mandal, P., Rawal, K., Ramaswamy, R., Bhattacharya, A. and Bhattacharya, S. "Identification of Insertion hot spots for non-LTR retrotransposons: Computational and Biochemical application to Entamoeba histolytica." Nucl. Acids Res., vol. 34, no. 20, pp. 5752-5763, 2006. (Lead author and equal contribution).
Dev, B.B., Malik A., Rawal, K., “Detecting motifs and patterns at mobile genetic element insertion site”. Bioinformation, vol. 8, pp.777-786, 2012.
Rawal, K., Dorji, S. Kumar, A., Ganguly, A. Grewal, A.S. “Identification and characterization of MGEs and their insertion sites in the gorilla genome”. Mobile Genetic Elements, vol.3, no.4, pp. e25675, 2012.
Rawal, K., Priya, A., Malik, A., Bahl, R., Ramaswamy, R., “Distribution of MGEs and their insertion sites in the Macaca mulatta genome”. Mobile Genetic Elements, vol.2, no.3, pp. 133-141, 2012
Bakre, A.A.,Rawal, K., Ramaswamy, R., Bhattacharya, A. and Bhattacharya, S., “The LINEs and SINEs of Entamoeba histolytica: Comparative analysis and genomic distribution.” Experimental Parasitology, vol. 110, no. 3, pp. 207-213, 2005.
Jagannadham, J., Jaiswal, H.K., Agrawal, S., Rawal, K., Comprehensive map of molecules implicated in obesity", PLoS ONE, vol. 11, no. 2 : e0146759. doi:10.1371/journal.pone.0146759, 2016.
Jagannadham, J., Jaiswal, H.K., Rawal, K., Deciphering relationships in disease networks using computational approaches: Fatty Liver, PCOD, Osteoarthritis, cholelithiasis & hyperlipdemia", International Journal of PharmTech Research, vol. 8, no. 1, pp. 127-134, 2015.
Jagannadham, J., Jaiswal, H.K., Agarwal, S., Rawal, K., Biomedical Text Mining of Obesity, Diabetes and hypertension genes. International Journal of Pharmaceutical Sciences Review and Research. vol 33(2), 182-186, 2015.
Agrawal, S., Rawal, K.,Sahu, A., Mahajan, S., Garg, P. and Bahl, R.,"To find gene distributions in PubMed abstracts using Perl software", Journal of Pharmacy Research 2013.
Jaiswal, H.K., Rawal, K., Jaganadham, J., Agrawal, S., “Evaluation of inhibition activity of Tetrahydrolipstatin analogues on Diacylglycerol lipase alpha using In-silico techniques”. Journal of Pharmacy Research, vol.5, no.6, pp. 3473-3477, 2012.
Rustagi Y, Jaiswal HK, Rawal, K., Kundu GC, Rani V (2015). Comparative Characterization of Cardiac Development Specific microRNAs: Fetal Regulators for Future. PLoS ONE.10(10): e0139359. https://doi.org/10.1371/journal.pone.0139359.
Gupta, R., Soni, Patnaik, Sood, I., Singh, R., Rawal, K., Rani, V. High AU content: A Signature of Upregulated miRNA in Cardiac Diseases, Bioinformation, vol. 3, pp.132-135, 2010, Vaccine Development.
Abbasi, B. A., Saraf, D., Sharma, T., Sinha, R., Singh, S., Gupta, P. Rawal, K. (2020, April 8). Identification of vaccine targets & design of vaccine against SARS-CoV-2 coronavirus using computational and deep learning-based approaches. https://doi.org/10.31219/osf.io/f8zyw
Kamal Rawal, v, Abbasi, B. A., et al (2020). Design of a multi-epitope Chagas disease vaccine by computational analysis of the Trypanosoma cruzi CL Brenner proteome. (Communicated)
Rawal k, Sinha R et al (2020) Vaxi – DL: A web-based Deep Learning (DL) server to identify Potential Vaccine Candidates, (Bioinformatics- communicated), COVID-19
Jethani B., Rawal, K. et al (2020). Clinical Characteristics and Remedy Profile of Patients with COVID-19: Retrospective Cohort Study, Accepted (In Press)
Rawal, K., Sinha R et al (2020) To Study the Effect of Unconventional Treatment Protocol on COVID-19 patients in Delhi using AI-based techniques (Lancet- communicated), Machine Learning
Rawal, K., Khurana T, Sharma H, . 2019. An extensive survey of molecular docking tools and their applications using text mining and deep curation strategies. PeerJ Preprints7:e27538v1 https://doi.org/10.7287/peerj.preprints.27538v1.
K. Rawal, P. Gaur and K. Kashive, LAP LAMBERT Academic Publishing GmbH & Co. KG, Saarbrücken Germany, October, 2014
Developing Networks in Obesity using Text Mining. K. Rawal, S. Agarwal and J. Jagannadham, LAP LAMBERT Academic Publishing GmbH & Co. KG, Saarbrücken Germany, September, 2014.
This study was conducted under the overall guidance of Dr. Kamal Rawal, who contributed to the protocol, critical evaluation of data, and manuscript. the pipeline was designed, constructed, and validated by Robin Sinha and Prashant Singh. the editing was done by Sweety Dattatraya Shinde and Ridhima.
I would like to express my deep gratitude to Amity University for delivering the administrative and technical assistance necessary for the completion of this study.
A Novel Artificial Intelligence and Machine learning-based scoring system for evaluating re-purposing potential of Valproic Acid drug in COVID-19.
Akanksha Aggarwal1, Prashant Singh2, Robin Sinha2, Preeti P2, Trapti Sharma2, Swarsat Kaushik Nath2, Kamal Rawal#1
1Delhi Institute of Pharmaceutical Sciences and Research, Delhi Pharmaceutical Sciences and Research University, New Delhi, 110017, India
2Amity Institute of Biotechnology, Amity University Uttar Pradesh, India.
#1Corresponding Author Email ID: kamal.rawal@gmail.com
Centre for Computational Biology and Bioinformatics, AIB Amity University, Noida
Background
Despite widespread vaccinations and the introduction of many repurposed medications, the rise of COVID-19 reinfection due to the SARS-CoV Omicron (B.1.1.529) variant has presented a significant challenge to health authorities across the world. There is a critical need for novel healthcare medications. Valproic acid (VPA) has been reported to be a beneficial drug due to its excellent property to hinder enveloped viral multiplication. Artificial Intelligence (AI) and Machine Learning (ML) have been used extensively to predict the repurposing potential of a drug by examining its past couple of years’ activity. The strategies for drug repositioning that will play a substantial role in this approach can be widely categorized into AI approaches, network-based models, and structure-based approaches.
Methodology
We have implemented a multimodal pipeline that operates on computational and machine learning strategies, namely molecular docking, molecular data, chemical information, clinical data, and gene expression to analyze the drugs for their sensitivity against covid-19. Here, we implemented the COV-DrugX pipeline to comprehend the physicochemical properties of VPA interactions with viral protein (targets), identify a place in the gene expression profile, additionally its potential role in human network databases.
Result and Conclusion
Here, we have analyzed the ML pipeline developed by our research group (Rawal et al., 2021) to identify the COVID-associated drug repurposing properties through AI and learning prediction module and use the COV-DrugX pipeline, that predicts the repurposing properties between different existing COVID-19 drugs. Nsp13 was found to have the highest binding affinity (-5.6 kcal/mol) with VPA. By using different modules, we have found that VPA displays physicochemical, biological, and other characteristics similar to existent COVID-19 drugs.
Keyword
Molecular Docking; Valproic Acid; VPA; Bioinformatics; drug repurposing; artificial intelligence; COVID-19; SAR-CoV-2
CoV-DrugX Pipeline | Home (kamalrawal. in) and Tools (kamalrawal. in)
At present, Coronavirus Disease 2019 (COVID-19) pandemic, caused by the Severe Acute Respiratory Syndrome Coronavirus-2 (SARS-CoV-2), has led to major oppression of worldwide healthcare infrastructure [Lai et al., 2020]. Widespread efforts are being put in by the scientists day in and day out to identify appropriate therapies against the viral disease. Although vaccine development and therapeutic antibodies are being industriously investigated, an increasing number of such promising approaches are the need of the hour, because they may lead us to continued efficacy in treating COVID-19 [Lu et al., 2020]. However, its inception has also resulted in a considerable number of clinical trials that assessed combinations of drugs made up of repurposed therapies ( references for drugs trials of VPA - NCT04513314 ).
VPA is a well-known histone deacetylases (HDAC2) inhibitor, that has received widespread acceptance for its usage in the treatment of epilepsy and other neurological illnesses during the last 50 years [Sébastien et al., 2010]. Recent data suggest that VPA may be useful in the treatment of some malignancies, due to its involvement in modifying epigenetic alterations by inhibiting histone deacetylases, which directly or indirectly influence the expression of genes related to the cell cycle, differentiation, and death [JE et al., 2020].VPA's direct impact on immune system cells has just lately been investigated. VPA affects the suppression of various activation mechanisms in different immune cells that contribute to an anti-inflammatory response is discussed in this study [Kong et al., 2020]. VPA displays its influences on the immune cells by demonstrating the expression of the cell cycle and death genes via epigenetic alterations [MedlinePlus Drug Information, n.d.]. VPA stimulates RNA interference, triggers histone methyltransferases, as well as restrains transcription factor activation along with blocking histone deacetylases [Li et al., 2020]. However, the effectiveness of VPA during the infectious phase is determined by the pathogen's biological characteristics and the related immune response; this is because VPA can support infection control or advancement [Soria-Castro et al.,2019]. VPA, because of its diverse effects, is a promising option for autoimmune diseases and hypersensitivity that is worth further investigation [Soria-Castro et al.,2019].
VPA has been shown to inhibit enveloped viral multiplication [Vazquez-Calvo A et al., 2011] and decreased levels of inflammatory cytokines in human monocyte-derived macrophages infected with Dengue virus e.g. tumor necrosis factor-alpha (TNF- alpha) and Interleukin-6 (IL-6) [Delgado F.G et al., 2018]. IL-6 levels were discovered to be a major biomarker of illness severity and death (illustrated in Figure 1) [Wang F et al., 2020]. VPA has been proven to alter macrophage responses to lipopolysaccharide (LPS) by decreasing IL-12p70 and TNF- production while raising IL-10 [Haskó et al., 2000]. Valproic acid can reduce angiotensin-converting enzyme 2(ACE2) and transmembrane serine protease 2 (TMPRSS2) production [MedlinePlus Drug Information, n.d.], all of which are essential for SARS-CoV-2 viral entry, ( see Figure 2) as well as supervise the immune cells and cytokine response to infection, minimizing end-organ damage [Pitt et al., 2021].
Previously, our research team has developed a number of machine learning and bioinformatics systems. Text mining and network biology-based systems are two examples [Jagannadham et al., 2016], likewise vaccine discovery system [Rawal et al. 2021], next-generation sequencing analysis system for cancer and other genomes [Preeti et al.,2021, Rawal et al 2011, Mandal et al 2006]. Moreover, they recently developed the first comprehensive pipeline and machine learning-based resource for analyzing mobile genomic elements and epigenetic mutations in different disorders [Rawal et al., 2011]. By merging large-scale experimental data with computational and comparative approaches, researchers were able to construct an unexplored methodology for determining the function of microRNAs in heart development and disorders. We discovered 353 previously unknown and 703 new miRNAs involved in heart maturation. Genes associated with the cell cycle, extracellular remodeling, apoptosis, metabolism, signaling pathways, transcriptional regulators, and chromatin remodeling seemed to be enriched in the target mRNAs [Rustagi et al., 2015], [Gupta, R., Soni, et al., 2010],[Abbasi, B. A. et al., 2020]
Many scholars have formed multiple new frameworks/portals for "obesity treatment", additional for diseases such as type 2 diabetes and hypertension amongst sedentary individuals by including important sections of the community over the use of social networks, data science systems, machine learning, sensors, and mobile apps. The technology is being used as a social service venture by the general population [K. Rawal, P. Gaur et al., 2014], for instance, the creation of a web-based weight-loss program[K. Rawal, S. Agarwal, et al., 2014]. Moreover, the advancement of the first molecular network on human obesity over screening >25 million PubMed records, for clinical studies, drug side effects, gene expression databases, and other information resources progressed the way for more computational bioinformatics research. Like the development of a machine learning model and a new text mining system for semi-automated screening of literature records with higher F scores [Jagannadham, J., Jaiswal, 2016].
We have designed a drug repurposing tool that examines the suitability of a drug (e.g. VPA) with existing FDA-approved drugs in COVID-19 treatment. Its algorithm is based on 14 different modules namely - CoV-DrugX, CoV-DrugX SE, CoV-DrugX Phenotype, CoV-DrugX Target, CoV-DrugX Condition, CoV-DrugX Symptoms, CoV-DrugX Gene Expressions, CoV-DrugX Pathway circuit, CoV-DrugX AI Ranking, CoV-DrugX Knowledge Graph, Drug-DL 11, Drug-DL 200, Drug Dock, Drug Dock Batch [Rawal et al., 2021].
Therefore, we run the pipeline to inspect the repurposing property for VPA. The docking module will give us the data on binding and interaction between COVID-19 proteins and the drug that we had given from the docking results we can determine whether the drug is suitable for repurposing or not.
For the examination, choose a particular drug, VPA, to check its re-purposing property. Further, we checked for its drug bank ID (we used DrugBank) and curated the isomeric smiles of it from the section chemical identifiers from the website. we saved the isomeric smiles in the .txt formats in the system. Next, we submitted its SMILES [CCCC(CCC)C(O)=O] as a query in the AI module (drugX pipeline), and one can choose some filters according to needs (see Figure 4 and Figure 5 ).
The results can be saved in CSV format and can be investigated accordingly (see Figure 6).
The second module (Table 2) suggests that the database includes data regarding drug COVID-19 repurposing 200 properties. For a more thorough assessment, all null values have been deleted from the acquired data source. Table 3 depicts the positive results that the given drug has shown interaction with various COVID-19 targets and has a number of the common targets too. Table 4 implies that the given drug has shown abnormal human gene expression during COVID-19. In response to the gene interaction, some of the drugs exhibited up-regulation (AKR1B7, HDAC7), though some of them showed down-regulation (HDAC2, CYP3A4), and in a couple of cases, a gene can be expressed either way depending on the factors unspecified (ALDH5A1, AKR1B1).
Another approach we used in our pipeline is docking-based predictions and we had built 3 docking-based modules where we have a human target-based docking module, viral proteins target-based module, and protein kinases associated with human proteins as targets for the query drugs.
When docked against 23 viral proteins, using drug_dock_viral, results are interpreted in the form of Binding Affinities (which shows the strength of binding interaction). The average binding affinity came out to be -4.30 Kcal/mol. Amongst all, the 5 highest interacting viral proteins were Nsp13, Nsp14, Npro, Nsp15, S_trimer with -5.6, -5.6, -5.1, -5.1, -5.0, respectively as their binding energies in kcal/mol. While, on the contrary, the lowest binding energy was observed in the case of Nsp1 and Orf10 with -2.6 and -3.2 Kcal/mol. See Table 5 and Figure 7 to understand data collected and a histogram showing the distribution of binding affinities against viral proteins. Table 6 and Figure 8 represents data for the docking results of VPA when docked against Human proteins (ACE2 and TMPRSS2)and associated protein kinases (GAK, JAK12, and AAK1) in form of binding affinities in kcal/mol.
We can conclude from Table 7 and Figure 9 that the drug has shown response with the SARS-CoV-2 pathway circuit. Table 8 reports that our database has data regarding having euclidean distance between drug and SARS-CoV-2 disease. Although, in a couple of cases, VPA showed COVID-19 side effects but doesn’t have any COVID-19 phenotype. However, it doesn’t cure any COVID-19 conditions.
We compiled the score of all 10 modules in the binary form of 0 and 1, except the docking modules where we found the results in this range of 0 and 1, taking into consideration the collective score for the docked proteins in the docking module (see Table 1). After analyzing VPA with the Cov-DrugX system, we've found 10 modules (out of 13 available) that had some information relevant to the drug, and out of which, 7 module reports a binary score of 1 (70%) and 3 modules reports a score of 0 (the modules Drug 11 properties, COVID-19 drug_Condition, and COVID-19 drug_phenotype) because of lacking any evident information about the drug in the source databases associated with these modules as there has been no gene expression associated to the drug reported yet. Hence, we can recognize the drug to be evaluated as a COVID-19 repurposable drug.
We can conclude from our analysis that VPA, a histone deacetylases (HDAC2) inhibitor has high capabilities to fight against the COVID-19 virus, and since filter studies had already proved that VPA and Nsp13 showed excellent binding affinity (-5.6 kcal/mol); should be able to treat COVID-19, is a promising alternative for COVID-19 repurposing drug treatment, as an effective, cheap, and easy to use drug which can hit all the targets. Clinical trials can help to find VPA’s role in governing the control of autoimmune diseases and hypersensitivity reactions that can be further explored, to discover more about the mechanism of action and characteristics features.
Lai CC, Shih TP, Ko WC, Tang HJ, Hsueh PR. Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and coronavirus disease-2019 (COVID-19): The epidemic and the challenges. International journal of antimicrobial agents. 2020 Mar 1;55(3):105924.
Wang H, Zheng X, Jin J, Zheng L, Guan T, Huo Y, Xie S, Wu Y, Chen W. LncRNA MALAT1 silencing protects against cerebral ischemia-reperfusion injury through miR-145 to regulate AQP4. Journal of biomedical science. 2020 Dec;27(1):1-2.
Kong FC, Ma CL, Zhong MK. Epigenetic effects mediated by antiepileptic drugs and their potential application. Current neuropharmacology. 2020 Feb 1;18(2):153-66.
Audia JE, Campbell RM. Histone modifications and cancer. Cold Spring Harbor perspectives in biology. 2016 Apr 1;8(4):a019521.
Chateauvieux S, Morceau F, Dicato M, Diederich M. Molecular and therapeutic potential and toxicity of valproic acid. Journal of Biomedicine and Biotechnology. 2010 Jul 29;2010.
Li G, Tian Y, Zhu WG. The roles of histone deacetylases and their inhibitors in cancer therapy. Frontiers in Cell and Developmental Biology. 2020 Sep 29;8:1004.
Valproic Acid: MedlinePlus Drug Information. Medlineplus.gov. Retrieved from National Library of Medicine
Soria-Castro R, Schcolnik-Cabrera A, Rodríguez-López G, Campillo-Navarro M, Puebla-Osorio N, Estrada-Parra S, Estrada-García I, Chacón-Salinas R, Chávez-Blanco AD. Exploring the drug repurposing versatility of valproic acid as a multifunctional regulator of innate and adaptive immune cells. Journal of immunology research. 2019 Mar 14;2019.
Vázquez-Calvo Á, Saiz JC, Sobrino F, Martín-Acebes MA. Inhibition of enveloped virus infection of cultured cells by valproic acid. Journal of virology. 2011 Feb 1;85(3):1267-74.
Delgado FG, Cárdenas P, Castellanos JE. Valproic acid downregulates cytokine expression in human macrophages infected with dengue virus. Diseases. 2018 Sep;6(3):59.
Wang F, Hou H, Luo Y, Tang G, Wu S, Huang M, Liu W, Zhu Y, Lin Q, Mao L, Fang M. The laboratory tests and host immunity of COVID-19 patients with different severity of illness. JCI insight. 2020 May 21;5(10).
Hasko G. Kuhel DG, Salzman AL, Szabo C. ATP suppression of interleukin-12 and tumor decreased necrosis factor-α release from macrophages. Br J Pharmacol. 2000;129:909-14.
Drugs, H. (2021). Valproic Acid: MedlinePlus Drug Information. Medlineplus.gov. Retrieved 13 December 2021, from https://medlineplus.gov/druginfo/meds/a682412.html.
Pitt B, Sutton NR, Wang Z, Goonewardena SN, Holinstat M. Potential repurposing of the HDAC inhibitor valproic acid for patients with COVID-19. European journal of pharmacology. 2021 Mar 2:173988.
CoV-DrugX Pipeline | Home. Drugx.kamalrawal.in. (2021). Retrieved 13 December 2021, from http://drugx.kamalrawal.in/drugx/
Rawal, K. and Ramaswamy, R., "Genome-wide analysis of mobile genetic elements insertion sites. Nucl. Acids Res.,vol. 39, no. 16, pp. 6864-6878, Sep. 2011.
Mandal, P., Rawal, K., Ramaswamy, R., Bhattacharya, A. and Bhattacharya, S. "Identification of Insertion hot spots for non-LTR retrotransposons: Computational and Biochemical application to Entamoeba histolytica." Nucl. Acids Res., vol. 34, no. 20, pp. 5752-5763, 2006. (Lead author and equal contribution).
Dev, B.B., Malik A., Rawal, K., “Detecting motifs and patterns at mobile genetic element insertion site”. Bioinformation, vol. 8, pp.777-786, 2012.
Rawal, K., Dorji, S. Kumar, A., Ganguly, A. Grewal, A.S. “Identification and characterization of MGEs and their insertion sites in the gorilla genome”. Mobile Genetic Elements, vol.3, no.4, pp. e25675, 2012.
Rawal, K., Priya, A., Malik, A., Bahl, R., Ramaswamy, R., “Distribution of MGEs and their insertion sites in the Macaca mulatta genome”. Mobile Genetic Elements, vol.2, no.3, pp. 133-141, 2012
Bakre, A.A.,Rawal, K., Ramaswamy, R., Bhattacharya, A. and Bhattacharya, S., “The LINEs and SINEs of Entamoeba histolytica: Comparative analysis and genomic distribution.” Experimental Parasitology, vol. 110, no. 3, pp. 207-213, 2005.
Jagannadham, J., Jaiswal, H.K., Agrawal, S., Rawal, K., Comprehensive map of molecules implicated in obesity", PLoS ONE, vol. 11, no. 2 : e0146759. doi:10.1371/journal.pone.0146759, 2016.
Jagannadham, J., Jaiswal, H.K., Rawal, K., Deciphering relationships in disease networks using computational approaches: Fatty Liver, PCOD, Osteoarthritis, cholelithiasis & hyperlipdemia", International Journal of PharmTech Research, vol. 8, no. 1, pp. 127-134, 2015.
Jagannadham, J., Jaiswal, H.K., Agarwal, S., Rawal, K., Biomedical Text Mining of Obesity, Diabetes and hypertension genes. International Journal of Pharmaceutical Sciences Review and Research. vol 33(2), 182-186, 2015.
Agrawal, S., Rawal, K.,Sahu, A., Mahajan, S., Garg, P. and Bahl, R.,"To find gene distributions in PubMed abstracts using Perl software", Journal of Pharmacy Research 2013.
Jaiswal, H.K., Rawal, K., Jaganadham, J., Agrawal, S., “Evaluation of inhibition activity of Tetrahydrolipstatin analogues on Diacylglycerol lipase alpha using In-silico techniques”. Journal of Pharmacy Research, vol.5, no.6, pp. 3473-3477, 2012.
Rustagi Y, Jaiswal HK, Rawal, K., Kundu GC, Rani V (2015). Comparative Characterization of Cardiac Development Specific microRNAs: Fetal Regulators for Future. PLoS ONE.10(10): e0139359. https://doi.org/10.1371/journal.pone.0139359.
Gupta, R., Soni, Patnaik, Sood, I., Singh, R., Rawal, K., Rani, V. High AU content: A Signature of Upregulated miRNA in Cardiac Diseases, Bioinformation, vol. 3, pp.132-135, 2010, Vaccine Development.
Abbasi, B. A., Saraf, D., Sharma, T., Sinha, R., Singh, S., Gupta, P. Rawal, K. (2020, April 8). Identification of vaccine targets & design of vaccine against SARS-CoV-2 coronavirus using computational and deep learning-based approaches. https://doi.org/10.31219/osf.io/f8zyw
Kamal Rawal, v, Abbasi, B. A., et al (2020). Design of a multi-epitope Chagas disease vaccine by computational analysis of the Trypanosoma cruzi CL Brenner proteome. (Communicated)
Rawal k, Sinha R et al (2020) Vaxi – DL: A web-based Deep Learning (DL) server to identify Potential Vaccine Candidates, (Bioinformatics- communicated), COVID-19
Jethani B., Rawal, K. et al (2020). Clinical Characteristics and Remedy Profile of Patients with COVID-19: Retrospective Cohort Study, Accepted (In Press)
Rawal, K., Sinha R et al (2020) To Study the Effect of Unconventional Treatment Protocol on COVID-19 patients in Delhi using AI-based techniques (Lancet- communicated), Machine Learning
Rawal, K., Khurana T, Sharma H, . 2019. An extensive survey of molecular docking tools and their applications using text mining and deep curation strategies. PeerJ Preprints7:e27538v1 https://doi.org/10.7287/peerj.preprints.27538v1.
K. Rawal, P. Gaur and K. Kashive, LAP LAMBERT Academic Publishing GmbH & Co. KG, Saarbrücken Germany, October, 2014
Developing Networks in Obesity using Text Mining. K. Rawal, S. Agarwal and J. Jagannadham, LAP LAMBERT Academic Publishing GmbH & Co. KG, Saarbrücken Germany, September, 2014.
This study was conducted under the overall guidance of Dr. Kamal Rawal, who contributed to the protocol, critical evaluation of data, and manuscript. the pipeline was designed, constructed, and validated by Robin Sinha and Prashant Singh. the editing was done by Sweety Dattatraya Shinde and Ridhima.
I would like to express my deep gratitude to Amity University for delivering the administrative and technical assistance necessary for the completion of this study.