1. Introduction
COVID-19 pandemic, caused by the SARS-CoV-2 virus, has created significant impacts throughout the entire world [1,2]. As of 5 November 2022, more than 632 million cases have been reported, resulting in about 6.5 million deaths globally [3]. Despite recent advances in drug discovery, the treatment of viral infections remains a significant challenge for scientists worldwide. Although various vaccines are now available on the market, they have not been sufficient to stop the spread of COVID-19, which is further complicated by variants of SARS-CoV-2 emerging continuously. An efficient and economic strategy for satisfying the urgent requirement for safe, effective, and affordable COVID-19 treatments is to repurpose the drugs already approved for other diseases [4,5]. A wide range of approaches is available for repurposing a drug, such as computational and bio-experimental approaches. With the use of various drug databases it is feasible to determine the biological function of a protein computationally and validate the process statistically. This technique, however, comes with its own set of challenges. One such challenge is data availability. Public access to high-quality, valuable information such as clinical trials data is still limited. Although there are a few open-access databases, such as the Drug Repurposing Hub [6], which provide drug repurposing candidates, very few are specific to COVID-19, such as Excelra [7] and NCATS [8].
Additionally, the significant sizes of these databases often create computational bottlenecks. Through this study we explored the problems of data accessibility, availability, and interpretability by developing CoviRx, a robust web-based application. The name CoviRx is a portmanteau of the terms COVID-19 and Rx (a medical prescription).
CoviRx provides users with an interactive interface to access 7817 drugs for COVID-19 drug repurposing. The web application (henceforth mentioned as either App or app) presents each drug’s physical and chemical properties, original indication, available data from multiple assays, COVID-19 clinical trials, and any red flags (pregnancy concerns and contraindications, among others) for these drugs. It calculates dynamically any drugs similar to the query drug using the Tanimoto coefficient [9], a popular metric used for finding similarity between molecules using their fingerprint descriptors. It displays the output along with the number of filters passed out of the total eleven designed by the research team. This computer science paper should be read in conjunction with research by MacRaild et al., 2022 [10], which describes the pharmacological methodology for systematic down-selection of repurposed drug candidates for COVID-19. To increase collaboration with the other researchers, features have been incorporated in CoviRx that allow registered users to submit their findings on any drug candidate to the CoviRx web app. Following a thorough peer review process, these submissions will be merged with the primary database, which we herein report. The user can download the available data as JSON (JavaScript Object Notation) or PDF (Portable Document Format) file, and unique URLs. This provides an elegant solution to the problem of data non-standardization and lack of data mining, interpretation, and manipulation-friendly datasets [11].
CoviRx has been designed according to current industry standards. Stringent checks are placed to ensure data consistency, security, and uniformity. Unique identifiers have been added for the individual drugs to ensure that they are easy to search and data for all the drugs are easily accessible. Modern frameworks such as Django [12,13], jQuery [14], and Bootstrap5 [15] have been used for development on a cloud hosting service (Azure) to ensure that the web app is scalable, i.e., able to process an increase in website traffic. It supports different user profiles to protect the integrity of data. Only registered users can submit to the website, to maintain data uniformity and reliability and prevent the malicious activities suffered by some prominent open-source platforms. Only the users with administration rights have modification privileges. Numerous graphical representations [16] are present throughout the web app to give the users an overview of the dataset and enhance their experience. Its modular architecture ensures that it is flexible enough to be extended to any other disease without disrupting its underlying functionality. CoviRx thus is a one-stop application for accessing high-quality data related to COVID-19 drug repurposing and, in the future, any other epidemic. CoviRx can be accessed at
2. Data Description
Drug data (with assays) were manually curated by our team from various research journals and publicly available drug databases. Regulatory approval data were obtained from the U.S. Food and Drug Administration (FDA) [17], Therapeutic Drug Administration, Australia (TGA) [18], and Inxight Drugs websites [19], while clinical trials data were acquired from
These drugs were then passed through a series of 11 filters, and the resultant drugs were screened for possible action against COVID-19 to narrow down to twelve drugs for further studies on COVID-19 drug repurposing [21].
Manual curation of data and datasets requires considerable time, labor, and financial resources. Therefore, to economise these resources, it was decided to open source this dataset and make it available with a user-friendly graphical interface so that other researchers working on their own filters for down-selecting drugs can use it as shown in Figure 1.
The CoviRx website consists of various identifiers such as synonyms, Chemical Abstract Service (CAS) number, and PubChem ID, ChEMBL ID, amongst others, which are hyperlinked to provide a seamless transition across different databases. Various drug-likeness properties such as molecular weight, Partition Coefficient (logP), hydrogen bond donors (HBD) and hydrogen bond acceptors (HBA) are also included for each compound to assist with early drug discovery programs. A drug’s original indication, mechanism of action (MOA), relevant pathways it interferes with, and associated targets are also included to provide a brief background about its original application. CoviRx provides users with SARS-CoV-2 drug repurposing data from several COVID-19 drug repurposing studies and the status of such drugs in COVID-19 trials. The website’s pharmacokinetics (PK) section includes information regarding a compound’s route of administration, the volume of distribution, clearance, protein binding, etc. The CoviRx website also consists of a specially curated section called “red flags”, which sheds light on drug–drug interactions, contraindications, pregnancy category of a drug, breastfeeding concerns, and severe side effects or black box warnings. To avoid early confounders in drug repurposing programs, information such as CAD (Cationic Amphiphilic Drugs) induced phospholipidosis and PAINS (Pan Assay Interference Compounds) are also included in the red flags section.
The data have been segmented into different tables for efficient storage. Figure 2 depicts different tables and the data fields used to store the drug data, and Figure 3 illustrates the same for data on users. The data types of each field and the type of constraints enforced (if any) are also highlighted.
3. Methods
3.1. Tech Stack and Design Principles
Modern technologies were employed to develop the front-end and back-end of the web app. Figure 4 demonstrates the interaction between different components like front-end and back-end servers, and the frameworks we have used to develop these. These frameworks are open-source, well documented, and highly flexible to use. Back-end servers, such as Apache [22] and cloud hosting platforms including Azure or AWS, were preferred over other alternatives as these provided support for both vertical and horizontal scalability. Currently, SQLite [23], a relational database management system, is being used to process website data; however, it will be migrated to PostgreSQL soon [24,25] as it offers better concurrency control [26]. Google OAuth [27] (login/logout), Google reCAPTCHA [28], and Google simple mail transfer protocol (SMTP) [29] have been used during development. CoviRx sends emails to its users for various purposes, such as a copy of responses filled by them in contact and ”contribute drug” forms; an invitation to the admin panel; and reporting errors that occur during drug upload. A python library called Premailer was used because emails do not support CSS or JS files, and it is important to deliver well-styled HTML emails to our users [30].
Few properties of database systems and websites [31,32,33] form the foundation for CoviRx’s architecture. These design principles have been discussed in detail in Table 1.
Figure 4Tech stack used for developing CoviRx and the interaction between these frameworks. User interacts with the front-end, built using HTML, CSS, JQuery3 [14], Bootstrap 5 [15], to request data. Front-end then interacts with the webservers, e.g., Azure, AWS, and Apache. These fetch data from the application servers powered by Django [12,13,34], which in turn interact with the database management system, i.e., SQLite [23]. Once these Data are extracted, it follows a reverse path back to the user.
[Figure omitted. See PDF]
3.2. CoviRx Architecture
An iterative software development model was followed [36] while designing CoviRx. First, this was achieved by creating a wireframe [37] for every front-end page, followed by a basic application implementation. The design components were then modified in each version by adding new features and functionalities. Each component was tested individually [38] before merging with the main framework. Upon its completion, the entire application was tested [39]. Figure 5 depicts various pages/features of CoviRx.
The CoviRx web app’s top hierarchy consists of the admin panel [40] and the user interface. It was a design decision to keep the user interface interactive and straightforward. Users can interact with our tool using features such as the search engine, drug overviews, contact forms, drug submission forms, etc. Features of the website’s home page are depicted in Figure 6.
The admin panel is for users with admin rights for the web app. It has two sections: accounts (associated with the users) and main. Admins can use the main section to contribute/upload drugs access website analytics data and custom fields. A feature to invite other users to the admin panel has been added, and admins can use it to send out invites with a seven-day expiry. Features of the admin panel are depicted in Figure 7.
Graphical representations have been used throughout the website to give the users a visual representation of the database, enhancing the user experience. CoviRx offers multi-language support [41] so that our open-source dataset is accessible to a broader range of users [42].
3.2.1. User Interface
The Search Engine
The standard and advanced search modes are two modes of usage, as depicted by Figure 8 and Figure 9, respectively. Both modes offer search-as-you-type [43] functionality. The significant difference between the modes is that only one identifier can be used to search for a drug in the case of normal mode. In the advanced search, multiple identifiers can be used simultaneously. The search results have been capped at five for the normal mode to reduce search time, and this limit is adjustable in the advanced search mode. The researchers can also choose a combination of different filters in the advanced search mode to restrict the drugs being displayed as per their requirement. The search results comprise drug metadata, i.e., drug synonyms, indication class, Simplified molecular-input line-entry system (SMILES) [44] representation, etc.
Drug Overview Page
This section of the website displays information about the query drug. The data have been divided into distinct categories like identifiers, drug-likeness, original indication, target models, clinical trials, red flags, filters passed, etc. The structure of the webpage is depicted in Figure 10. Specific data points related to drug activity on the original target have not been included; they do not have well-defined values or data uniformity but can be added in the future. Data points such as original indications and red flags have been added to aid researchers. In contrast, other fields like pharmacokinetics (PK) would be helpful to individuals who wish to model their filters and down-select drugs. Visual representation of the filters used to down-select drugs and the filters passed by each drug has been added for those who would like to investigate this further. Drugs are ranked based on activity scores calculated using the assay data as described in MacRaild et al., 2022 [10].
Another important feature of this page is the dynamic calculation and display of similar drug candidates. Drug similarity is calculated using the Tanimoto coefficient [9,45], which is the ratio of common fingerprints and the number of total fingerprints. The RDKit python library [46] was used to generate the fingerprints of the drugs and calculate the Tanimoto similarity between them [47]. The threshold for similarity is kept at a value of 0.7, less than the suggested value of 0.85 [48], to increase the number of similar drugs. Once the similarity is calculated, the page displays the top 5 (i.e., highest five Tanimoto coefficients) candidates.
Contribute Drug
CoviRx supports and encourages the community to help us enlarge this open-source database. Authenticated individuals can contribute to the web app using the “contribute drug” feature, an overview of which is depicted in Figure 11. Only registered users can make these contributions to ensure data integrity and uniformity. All contributed drugs would be verified manually by a team of admin users to ensure data authenticity and reliability.
Contact
Using the contact page, users can contact administrators to give any feedback or to request access to contribute drugs. A reCAPTCHA [28,49] check has been added to prevent robots from submitting the form.
3.2.2. Admin Panel
Bulk Upload Support
To facilitate the process of bulk drug upload, CoviRx has a feature which enables the admin users to upload large amounts of drug data in one go by directly submitting a CSV file. A single drug upload/delete/modify functionality is also available. CoviRx evaluates every drug against multiple constraints before adding it to the primary database. These constraints ensure the validity of the drugs, so drugs that fail them are considered “invalidated” and shared with the admin users to reverify the data. Figure 12 depicts the various sub-features of the bulk upload feature.
Website Analytics
Mechanisms have been developed to measure the website traffic, the details of which are available in the admin panel. According to the website load, this was added to plan the resources for hosting the web application. It will also help track the trends about the most popular drugs in the database. While monitoring the website traffic, only a session ID cookie [50] is used, and no other details like IP address are saved to ensure users’ privacy [51].
Custom Fields
Every data field of a drug displayed on the CoviRx platform maps to an attribute in the database. This implies that the code needs to be altered if a field is added, deleted, or modified for a particular drug. Custom fields have been added to overcome this problem of repeatedly changing the code and are critical components in the admin panel. It is a JSON field in the database that stores key-value pairs, with the key holding the name of the drug field and the value holding the value for that field [52]. This feature would make CoviRx extensible and flexible in the long run.
3.3. Articles Scraping
To ensure CoviRx does not end up being a static database after its creation, we have introduced a new feature to keep the database up to date with new articles published by peer-reviewed journals and reputed preprint servers such as bioRxiv and medRxiv. CoviRx looks up for new articles periodically (at present every month) to find new assay data for drugs present in the CoviRx database. The papers found are then assigned to admin users for verification. We selected Google Scholar [53] as a platform to find new articles as it covers a very wide range of publications including preprints. A paper can be found at multiple places (preprint server, peer reviewed journal, etc.), in different formats (html, pdf, etc.), therefore CoviRx ensures that only the most recent version gets scraped to prevent duplicate entries of the same article.
The scraped articles are then equitably assigned to admin users, who, as domain experts, verify the article and update the database. When an admin user becomes inactive or leaves, the articles assigned to them are distributed amongst the remaining admin users automatically. Articles are grouped drug-wise to ease verification as shown in Figure 13. The admin users can then view a list of the articles scraped, along with the drug, target model associated, and other keywords used for scraping as shown in Figure 14. If the articles found are relevant, the drug’s COVIDvid-19 assay data can be updated with the help of the “Update form” as shown in Figure 15.
Since CoviRx crowd-sources the article verification job to pharmacy students, it becomes imperative to monitor the articles being verified, as shown in Figure 16. CoviRx also allows admin superusers (the highest administration privilege) to revert the changes made by admin users so that any incorrect change can be undone, as shown in Figure 17. In the future, the manual verification process of scraped articles can be automated using NLP techniques [54].
4. User Notes
The supplementary information provides additional information to the users. The external libraries and external APIs used are listed in Tables S1 and S2, respectively. A few more screenshots of the user interface, admin panel and the formulas used are provided in Figure S3, Figure S4 and Text S5, respectively. Additional information for contributing drugs and privacy policy of CoviRx is covered in Text S6, and URL S7 includes a link to the GitHub repository for CoviRx; URL S8 links to more detailed technical documentation, and Text S10 provides information on reusability for developers. All the API endpoints provided by CoviRx are covered in detail, with example in Text S9.
5. Conclusions
CoviRx provides users with repurposed drugs that could have anti-viral activity against SARS-CoV-2, based on pharmacological information, assay activity, and drug status in COVID-19 clinical trials. Its user-friendly interface and multi-language support improve data accessibility. Various drug-likeness properties such as molecular weight, logP, hydrogen bond donors (HBD), and hydrogen bond acceptors (HBA), amongst others, are included for each compound to assist with early drug discovery programs. The users who wish to design their own filters for drug down-selection can benefit from the website’s pharmacokinetics (PK) section, which provides information regarding a compound’s route of administration, the volume of distribution, clearance, protein binding, etc. This platform also encourages its users to help expand this open-source dataset by submitting their drug repurposing data, and to register as volunteer administrators to keep the database error free and up to date. CoviRx has a wide variety of applications, and thus will help advance the field of drug repurposing for the current pandemic as well as future epidemics.
Conceptualization, H.A.J., A.K., M.M.S., N.L.T., D.J.C., V.A. and S.S.V.; methodology, H.A.J., and V.A.; software, H.A.J. and V.A.; validation, C.B., A.K., F., M.-U.-R.M., S.M., A.V.K., C.A.M., I.K.S., C.M.J.K., R.M.S., E.A.P., N.L.T., D.J.C. and S.S.V.; formal analysis, H.A.J.; investigation, H.A.J.; resources, S.M., V.A. and S.S.V.; data curation, A.K., F., M.-U.-R.M., S.M., M.M.S., C.A.M., I.K.S., A.L.P., C.M.J.K., N.L.T. and D.J.C.; writing—original draft preparation, H.A.J., C.B., V.A. and S.S.V.; writing—review and editing, H.A.J., V.A., C.B., A.K., F., M.-U.-R.M., S.M., M.M.S., A.V.K., R.C., C.A.M., I.K.S., A.L.P., M.A.C., C.M.J.K., R.M.S., E.A.P., N.L.T., D.J.C. and S.S.V.; visualization, H.A.J. and V.A.; supervision, A.K., D.J.C., V.A. and S.S.V.; project administration, S.M., I.K.S., V.A. and S.S.V.; funding acquisition, S.S.V. All authors have read and agreed to the published version of the manuscript.
Not applicable.
Not applicable.
Any underlying data not presented can be provided by the corresponding author upon reasonable request.
The authors would like to acknowledge the contributions made to the project by the broader sySTEMs Initiative: Aditya V Vashi, Alexander J McAuley, Carmel M O’Brien, Carol Lee, Chandana Venkatadri, Darcie Cooper, David Beale, Eugene Athan, Elizabeth Vincan, James Hudson, Jenny O’Connell, Jian-Wei Liu, John Noel Viana, Kanta Subbarao, Kathie Burkett, Kim R Blasdell, Laurence Wilson, Lee Trinidad, Mary Tachedjian, Matthew P Bruce, Michael Kuiper, Nagendrakumar Balasubramanian Singanallur, Nathan J Gödde, Petrus Jansen van Vuren, Sarah Goldie, Shane Riddell, Simone Clayton, Simran Chahal, Stephanie Keating, Trevor Drew and Vinod Sundaramoorthy.
The authors declare no conflict of interest.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Enlarge this image.Figure 13. Articles are clubbed drugwise, and 5 drugs are displayed on every page to ease verification.
Enlarge this image.Figure 14. The URL, target model and drug associated with every article are listed.
Enlarge this image.Figure 15. The update form can be used by the verifier to update or add new target model data based on the evidence presented in the respective journal article.
Enlarge this image.Figure 16. The articles verified by the pharmacy students can be tracked by the admin superusers.
Enlarge this image.Figure 17. Feature to monitor the changes made by every user in detail and revert them if needed.
Design Principles adopted during the development of CoviRx.
|
|
|
|
|
|---|---|---|---|
| Access Management | To ensure that no unauthorized access happens, users were divided into three broad categories:
|
|
--- |
| Data Security | Protection against cross-site scripting (XSS), cross-site request forgery (CSRF), SQL injection, and clickjacking is provided by Django. Secure communication between the web app and the server is established as only encrypted Data are shared via the HTTPS protocol. |
|
Django, HTTPS Protocols |
| Scalability | Modern cloud hosting providers and back-end servers support both vertical and horizontal scaling. |
|
Django, Apache back-end server, AWS, Azure, Google Cloud Platform (GCP) |
| Modularity | The entire website has been broken down into multiple independent components which are interchangeable. The database components are well encapsulated from program components of the web app. | It adds a layer of abstraction that provides data program independence.
|
Django |
| Integrity Constraints | Various constraints are in place to ensure that only data that pass all quality checks are merged with the primary database. Unique identifier constraint is enforced in multiple fields like drug name, ID, etc., Char fields with maximum length constraints over text fields are used to optimize memory usage. |
|
--- |
| Auto Versioning of Static Files | A string specifying the version of the static file is appended to the URL pointing to the static files. This version is updated, forcing the browser to load the latest version instead of the one from the cache. |
|
--- |
| Flexibility and Responsiveness | The web app is responsive to a wide range of screen sizes. Django template language makes designing dynamic front-end components easy. |
|
Django, Bootstrap5, jQuery3, jQuery3 UI |
| Backup and Recovery | Data backups are done weekly. | Protects the data from system failure, both hardware and software. | --- |
| Concurrency Control | SQLite is not optimized for performing concurrent operations and often results in Database is locked exceptions. So, we intend to shift to PostgreSQL in the future so that simultaneous operations do not lead to abnormalities. |
|
SQLite (shifting to PostgreSQL in future) |
| Data Persistency | If an exception happens between a transaction (write operation), the data to be written might get destroyed. We plan to protect our system in rare cases of failure like this to ensure durability. | Data is never at risk. | Celery [ |
Supplementary Materials
The following supporting information can be downloaded at:
References
1. Zhu, N.; Zhang, D.; Wang, W.; Li, X.; Yang, B.; Song, J.; Zhao, X.; Huang, B.; Shi, W.; Lu, R. et al. A Novel Coronavirus from Patients with Pneumonia in China, 2019. N. Engl. J. Med.; 2020; 382, pp. 727-733. [DOI: https://dx.doi.org/10.1056/NEJMoa2001017] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31978945]
2. Huang, C.; Wang, Y.; Li, X.; Ren, L.; Zhao, J.; Hu, Y.; Zhang, L.; Fan, G.; Xu, J.; Gu, X. et al. Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China. Lancet; 2020; 395, pp. 497-506. [DOI: https://dx.doi.org/10.1016/S0140-6736(20)30183-5]
3. Dong, E.; Du, H.; Gardner, L. An interactive web-based dashboard to track COVID-19 in real time. Lancet Infect. Dis.; 2020; 20, pp. 533-534. [DOI: https://dx.doi.org/10.1016/S1473-3099(20)30120-1]
4. Wang, X.; Guan, Y. COVID-19 drug repurposing: A review of computational screening methods, clinical trials, and protein interaction assays. Med. Res. Rev.; 2021; 41, pp. 5-28. [DOI: https://dx.doi.org/10.1002/med.21728] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32864815]
5. Pushpakom, S.; Iorio, F.; Eyers, P.A.; Escott, K.J.; Hopper, S.; Wells, A.; Doig, A.; Guilliams, T.; Latimer, J.; McNamee, C. et al. Drug repurposing: Progress, challenges and recommendations. Nat. Rev. Drug Discov.; 2018; 18, pp. 41-58. [DOI: https://dx.doi.org/10.1038/nrd.2018.168] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30310233]
6. Corsello, S.M.; Bittker, J.A.; Liu, Z.; Gould, J.; McCarren, P.; Hirschman, J.E.; Johnston, S.E.; Vrcic, A.; Wong, B.; Khan, M. et al. The Drug Repurposing Hub: A next-generation drug library and information resource. Nat. Med.; 2017; 23, pp. 405-408. [DOI: https://dx.doi.org/10.1038/nm.4306] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28388612]
7. Excelra | Covid-19-Drug-Repurposing-Database. Available online: https://www.excelra.com/covid-19-drug-repurposing-database/ (accessed on 31 January 2022).
8. Brimacombe, K.R.; Zhao, T.; Eastman, R.T.; Hu, X.; Wang, K.; Backus, M.; Baljinnyam, B.; Chen, C.Z.; Chen, L.; Eicher, T. et al. An OpenData portal to share COVID-19 drug repurposing data in real time. BioRxiv; 2020; [DOI: https://dx.doi.org/10.1101/2020.06.04.135046]
9. Tanimoto, T.T. Elementary Mathematical Theory of Classification and Prediction; International Business Machines Corporation: New York, USA, 1958.
10. MacRaild, C.A.; Mohammed, M.-U.-R.; Faheem,; Murugesan, S.; Styles, I.K.; Peterson, A.L.; Kirkpatrick, C.M.J.; Cooper, M.A.; Palombo, E.A.; Simpson, M.M. et al. Systematic Down-Selection of Repurposed Drug Candidates for COVID-19. Int. J. Mol. Sci.; 2022; 23, 11851. [DOI: https://dx.doi.org/10.3390/ijms231911851] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36233149]
11. Talevi, A.; Bellera, C.L. Challenges and opportunities with drug repurposing: Finding strategies to find alternative uses of therapeutics. Expert Opin. Drug Discov.; 2019; 15, pp. 397-401. [DOI: https://dx.doi.org/10.1080/17460441.2020.1704729] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31847616]
12. Django (Version 3.2). Available online: https://www.djangoproject.com/ (accessed on 26 January 2022).
13. Holovaty, A.; Kaplan-Moss, J. The Definitive Guide to Django: Web Development Done Right; Apress, Berkely, CA, USA, 2009.
14. Official jQuery Blog. Available online: https://blog.jquery.com/2021/03/02/jquery-3-6-0-released/ (accessed on 26 January 2022).
15. Bootstrap 5: Documentation. Available online: https://getbootstrap.com/docs/5.0/getting-started/introduction/ (accessed on 26 January 2022).
16. Charts | Google Developers. Available online: https://developers.google.com/chart (accessed on 7 February 2022).
17. Drugs@FDA: FDA-Approved Drugs. Available online: https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm (accessed on 26 January 2022).
18. TGA Search. Available online: https://tga-search.clients.funnelback.com/s/search.html?query=&collection=tga-artg (accessed on 26 January 2022).
19. Inxight Drugs. Available online: https://drugs.ncats.io/ (accessed on 26 January 2022).
20. Home—ClinicalTrials.gov. Available online: https://clinicaltrials.gov/ (accessed on 26 January 2022).
21. McAuley, A.J.; Jansen van Vuren, P.; Mohammed, M.-U.-R.; Faheem,; Goldie, S.; Riddell, S.; Gödde, N.J.; Styles, I.K.; Bruce, M.P.; Chahal, S. et al. Use of Human Lung Tissue Models for Screening of Drugs against SARS-CoV-2 Infection. Viruses; 2022; 14, 2417. [DOI: https://dx.doi.org/10.3390/v14112417]
22. Fielding, R.T.; Kaiser, G. Collaborative work: The apache http server project. IEEE Internet Comput.; 1997; 1, pp. 88-90. [DOI: https://dx.doi.org/10.1109/4236.612229]
23. Owens, M. The Definitive Guide to SQLite; Apress, Berkely, CA, USA, 2006.
24. PostgreSQL Documentation. Available online: https://www.postgresql.org/docs/current/index.html (accessed on 9 November 2022).
25. Douglas, K.; Douglas, S. PostgreSQL: A Comprehensive Guide to Building, Programming, and Administering PostgresSQL Databases; SAMS Publishing: Carmel, IN, USA, 2003.
26. SQLite vs. MySQL vs. PostgreSQL: A Comparison Of Relational Database Management Systems | DigitalOcean. Available online: https://www.digitalocean.com/community/tutorials/sqlite-vs-mysql-vs-postgresql-a-comparison-of-relational-database-management-systems (accessed on 26 January 2022).
27. Using OAuth 2.0 to Access Google APIs. Available online: https://developers.google.com/identity/protocols/oauth2 (accessed on 26 January 2022).
28. reCAPTCHA. Available online: https://www.google.com/recaptcha/about/ (accessed on 26 January 2022).
29. IMAP, POP, and SMTP | Gmail IMAP | Google Developers. Available online: https://developers.google.com/gmail/imap/imap-smtp (accessed on 26 January 2022).
30. GitHub—Peterbe/Premailer: Turns CSS Blocks into Style Attributes. Available online: https://github.com/peterbe/premailer (accessed on 26 January 2022).
31. Ramakrishnan, R.; Gehrke, J.; Gehrke, J. Database Management Systems; McGraw-Hill: New York, NY, USA, 2003; Volume 3.
32. Silberschatz, A.; Korth, H.F.; Sudarshan, S. Database System Concepts; 7th ed. McGraw-Hill: New York, NY, USA, 2019.
33. Elmasri, R.; Navathe, S.B.; Elmasri, R.; Navathe, S.B. Fundamentals of Database Systems; Springer: Berlin/Heidelberg, Germany, 2000.
34. Django Template Language. Available online: https://docs.djangoproject.com/en/3.2/topics/templates/ (accessed on 26 January 2022).
35. celery. Available online: https://pypi.org/project/celery/ (accessed on 7 February 2022).
36. Larman, C.; Basili, V.R. Iterative and incremental development: A brief history. Computer; 2003; 36, pp. 47-56. [DOI: https://dx.doi.org/10.1109/MC.2003.1204375]
37. Becker, S.A.; Berkemeyer, A. Rapid application design and testing of Web usability. IEEE Multimed.; 2002; 9, pp. 38-46. [DOI: https://dx.doi.org/10.1109/MMUL.2002.1041947]
38. Weyuker, E.J. Testing component-based software: A cautionary tale. IEEE Softw.; 1998; 15, pp. 54-59. [DOI: https://dx.doi.org/10.1109/52.714817]
39. Pezze, M.; Young, M. Software Testing and Analysis: Process, Principles, and Techniques; John Wiley & Sons: Hoboken, NJ, USA, 2008.
40. django-admin-interface. Available online: https://pypi.org/project/django-admin-interface/ (accessed on 7 February 2022).
41. Cloud Translation documentation | Google Cloud. Available online: https://cloud.google.com/translate/docs (accessed on 26 January 2022).
42. Melitz, J.; Toubal, F. Native language, spoken language, translation and trade. J. Int. Econ.; 2014; 93, pp. 351-363. [DOI: https://dx.doi.org/10.1016/j.jinteco.2014.04.004]
43. Ji, S.; Li, G.; Li, C.; Feng, J. Efficient interactive fuzzy keyword search. Proceedings of the WWW’09—18th International Conference on World Wide Web; Madrid Spain, 20–24 April 2009; pp. 371-380. [DOI: https://dx.doi.org/10.1145/1526709.1526760]
44. Weininger, D. SMILES, a Chemical Language and Information System: 1: Introduction to Methodology and Encoding Rules. J. Chem. Inf. Comput. Sci.; 1988; 28, pp. 31-36. [DOI: https://dx.doi.org/10.1021/ci00057a005]
45. Bajusz, D.; Rácz, A.; Héberger, K. Why is Tanimoto index an appropriate choice for fingerprint-based similarity calculations?. J. Cheminformatics; 2015; 7, pp. 1-13. [DOI: https://dx.doi.org/10.1186/s13321-015-0069-3] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26052348]
46. RDKit. Available online: https://www.rdkit.org/ (accessed on 26 January 2022).
47. Willett, P.; Barnard, J.M.; Downs, G.M. Chemical similarity searching. J. Chem. Inf. Comput. Sci.; 1998; 38, pp. 983-996. [DOI: https://dx.doi.org/10.1021/ci9800211]
48. Two Similar Compounds with a Low (Smaller Than 0.85) Tanimoto Coefficient?. Available online: https://www.researchgate.net/post/Two_similar_compounds_with_a_low_smaller_than_085_Tanimoto_coefficient2 (accessed on 26 January 2022).
49. Zhang, Y.; Gao, H.; Pei, G.; Luo, S.; Chang, G.; Cheng, N. A survey of research on CAPTCHA designing and breaking techniques. Proceedings of the 2019 18th IEEE International Conference on Trust, Security and Privacy in Computing and Communications/13th IEEE International Conference on Big Data Science and Engineering, TrustCom/BigDataSE 2019; Rotorua, New Zealand, 5–8 August 2019; pp. 75-84. [DOI: https://dx.doi.org/10.1109/TrustCom/BigDataSE.2019.00020]
50. Lacroix, K.; Loo, Y.L.; Choi, Y.B. Cookies and Sessions: A Study of What They Are, How They Work and How They Can Be Stolen. Proceedings of the 2017 International Conference on Software Security and Assurance, ICSSA 2017; Altoona, PA, USA, 24–25 July 2017; pp. 20-24. [DOI: https://dx.doi.org/10.1109/ICSSA.2017.9]
51. Chung, W.; Paynter, J. Privacy issues on the Internet. Proceedings of the 35th Annual Hawaii International Conference on System Sciences; Big Island, HI, USA, 10 January 2002; [DOI: https://dx.doi.org/10.1109/HICSS.2002.994191]
52. django-flat-json-widget · PyPI. Available online: https://pypi.org/project/django-flat-json-widget/ (accessed on 7 February 2022).
53. Google Scholar. Available online: https://scholar.google.com (accessed on 12 October 2022).
54. Subramanian, S.; Baldini, I.; Ravichandran, S.; Katz-Rogozhnikov, D.A.; Ramamurthy, K.N.; Sattigeri, P.; Varshney, K.R.; Wang, A.; Mangalath, P.; Kleiman, L.B. Drug Repurposing for Cancer: An NLP Approach to Identify Low-Cost Therapies. arXiv; 2019; arXiv: 1911.07819
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; Agarwal, Vinti 2 ; Bansal, Chaarvi 3 ; Kumar, Anupama 4 ; Faheem 5
; Muzaffar-Ur-Rehman, Mohammed 6
; Sankaranarayanan Murugesan 6
; Simpson, Moana M 7 ; Karpe, Avinash V 8
; Chandra, Rohitash 9
; MacRaild, Christopher A 10
; Styles, Ian K 10 ; Peterson, Amanda L 11 ; Cooper, Matthew A 12
; Kirkpatrick, Carl M J 13 ; Shah, Rohan M 14
; Palombo, Enzo A 14
; Trevaskis, Natalie L 10
; Creek, Darren J 10
; Vasan, Seshadri S 15
1 Birla Institute of Technology and Science, Department of Electrical and Electronics Engineering, Pilani 333031, Rajasthan, India
2 Birla Institute of Technology and Science, Department of Computer Science and Information Systems, Pilani 333031, Rajasthan, India
3 Birla Institute of Technology and Science, Department of Computer Science and Information Systems, Pilani 333031, Rajasthan, India; School of Mathematics and Statistics, University of New South Wales, Sydney, NSW 1466, Australia
4 Commonwealth Scientific and Industrial Research Organisation, Land and Water, Waite Campus, Urrbrae, SA 5064, Australia
5 Birla Institute of Technology and Science, Department of Pharmacy, Pilani 333031, Rajasthan, India; Department of Medicinal Chemistry, University of Utah, Salt Lake City, UT 84112, USA
6 Birla Institute of Technology and Science, Department of Pharmacy, Pilani 333031, Rajasthan, India
7 Compounds Australia, Griffith Institute for Drug Discovery, Griffith University, Nathan, QLD 4111, Australia
8 Department of Chemistry and Biotechnology, Swinburne University of Technology, Hawthorn, VIC 3122, Australia; Commonwealth Scientific and Industrial Research Organisation, Land and Water, Dutton Park, QLD 4102, Australia
9 School of Mathematics and Statistics, University of New South Wales, Sydney, NSW 1466, Australia
10 Drug Delivery, Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC 3800, Australia
11 Drug Delivery, Disposition and Dynamics, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC 3800, Australia; Bio21 Institute, University of Melbourne, Parkville, VIC 3052, Australia
12 Institute for Molecular Bioscience, The University of Queensland, Brisbane, QLD 4072, Australia
13 Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, Parkville, VIC 3052, Australia
14 Department of Chemistry and Biotechnology, Swinburne University of Technology, Hawthorn, VIC 3122, Australia
15 Commonwealth Scientific and Industrial Research Organisation, Australian Centre for Disease Preparedness, Geelong, VIC 3220, Australia; Department of Health, 189 Royal Street, East Perth, WA 6004, Australia; Department of Health Sciences, University of York, York YO10 5DD, UK