Abstract

The drug discovery and development process is very lengthy, highly expensive, and extremely complex in nature. Considering the time and cost constrictions associated with conventional drug discovery, new methods must be found to enhance the declining efficiency of traditional approaches. AI is transformation in drug discovery, increasing interest from investors, industrial and academic scientists. Optimizing properties related to pharmacodynamics, pharmacokinetics, and clinical outcomes are required for Successful drug discovery. This review discusses about the drug discovery and development process, implementation and role of AI tools in drug development and challenges and future of AI. The combination of AI and drug discovery offers a promising strategy to overcome the challenges and complexities of the pharmaceutical industry. New research tools may need to be evaluated to explore new targets. The journey from initial discovery to marketed drug is a long and difficult task. From discovery to approved drug takes about 12 to 15 years and requires an investment of about $1 billion. On average, one million molecules are screened, but only one will be investigated in late-stage clinical trials and ultimately delivered to patients.

Keywords

Artificial intelligence, Machine learning, Computer-aided drug discovery, Drug discovery and development.

Introduction

       
           
       
1. Discovery: The discovery stage starts with research and development in a laboratory setting. Researchers identify target molecules such as genes or proteins that are believed to have the potential to contribute to the development or aggravation of a disease. 

       
           
       
Role of artificial intelligence (AI) in drug discovery. AI's applications in drug discovery are revolutionizing the field by enhancing various stages of the process, making it faster, more efficient, and cost-effective. Here’s a detailed look at how AI is being applied across different facets of drug discovery:

 

---

Challenges and Limitations of AI in Drug Discovery

The similarity-based methods employed in these technologies typically do not account for heterogeneous information defined in a relationship network. Combining feature-based and similarity-based methods can help overcome this restriction. The fact that AI/MLbased models need a lot of training and that the requirements vary depending on the application is another drawback. Due to the learning qualities ingrained in the modelling data, DL models produce “black boxes” that are challenging to interpret; however, their use is growing in popularity. Nevertheless, their use is limited by the need for large amounts of high-quality data that are expensive to produce and frequently kept private. The use of inadequate quantities of high-quality data also affects the performance and dependability of DL models.

While AI holds great promise in drug discovery, it also faces several challenges and limitations that can impact its effectiveness and integration into the drug development process. Here are some key challenges:

1. Data Quality and Availability

• Data Quality: AI models are heavily dependent on the quality of data they are trained on. Poor-quality, incomplete, or biased data can lead to inaccurate predictions and unreliable results.
• Data Scarcity: In some areas, high-quality datasets are limited, particularly for rare diseases or novel drug targets. This scarcity can hinder the development and training of robust AI models.

2. Interpreting AI Models

• Black Box Problem: Many AI models, especially deep learning algorithms, operate as "black boxes," making it difficult to understand how they arrive at specific conclusions. This lack of transparency can be problematic for scientific validation and regulatory approval.
• Complexity: The complexity of AI models can make it challenging to interpret their outputs, which is critical for understanding the underlying biological mechanisms and validating predictions.

3. Generalization and Overfitting

• Overfitting: AI models trained on specific datasets may perform well on those datasets but fail to generalize to new, unseen data. Overfitting occurs when the model becomes too tailored to the training data and loses its ability to predict accurately on other datasets.
• Bias: AI models can inherit biases present in the training data, leading to skewed results and potential disparities in drug efficacy or safety across different populations.

4. Integration with Existing Workflows

• Workflow Integration: Integrating AI tools into traditional drug discovery workflows can be complex and requires collaboration between data scientists, biologists, and chemists. Adapting existing processes and infrastructure to accommodate AI-driven approaches can be challenging.
• Resistance to Change: There may be resistance from researchers and industry professionals who are accustomed to traditional methods and may be hesitant to adopt new AI-driven approaches.

5. Regulatory and Ethical Issues

• Regulatory Approval: The use of AI in drug discovery raises regulatory challenges, including ensuring that AI-generated findings meet the rigorous standards for safety and efficacy required for drug approval.
• Ethical Concerns: AI applications must address ethical issues related to data privacy, consent, and the potential for unintended consequences. Ensuring responsible use of AI in drug discovery is crucial.

6. Validation and Reproducibility

• Model Validation: Validating AI models requires rigorous testing and independent verification to ensure that predictions are accurate and reliable. This validation process can be resource-intensive and time-consuming.
• Reproducibility: Ensuring that AI models produce consistent results across different studies and datasets is essential for scientific reproducibility. Variability in results can undermine confidence in AI-driven discoveries.

7. Complexity of Biological Systems

• Biological Complexity: Biological systems are highly complex and not fully understood. AI models may struggle to capture the full complexity of biological interactions and pathways, leading to incomplete or inaccurate predictions.
• Dynamic Nature: Biological systems are dynamic and can change over time, which can affect the accuracy of AI predictions if the models are not updated to reflect these changes.

8. Computational and Resource Constraints

• Resource Intensiveness: Training and deploying advanced AI models require significant computational resources and infrastructure, which can be costly and may limit access for smaller organizations or research groups.
• Scalability: Scaling AI solutions from research settings to large-scale drug discovery projects can be challenging and may require substantial adjustments and optimizations.

9. Interdisciplinary Collaboration

• Skill Gaps: Successful implementation of AI in drug discovery requires interdisciplinary collaboration, including expertise in data science, biology, chemistry, and pharmacology. Bridging these diverse skill sets can be difficult and may require additional training and coordination.

10. Ethical AI Development

• Bias and Fairness: Ensuring that AI systems are developed and trained in a way that minimizes bias and promotes fairness is critical. Addressing potential biases in training data and model outcomes is an ongoing challenge.

Despite these challenges, ongoing research and development efforts aim to address these limitations and improve the integration of AI in drug discovery. As technology evolves and best practices are established, many of these issues may be mitigated, leading to more effective and reliable AI-driven drug discovery processes.

Future Perspective (23)

The future of AI in drug discovery and development is promising and transformative. Here are some key perspectives:

1. Enhanced Target Identification: AI algorithms can analyze vast datasets, including genomic, proteomic, and metabolomic information, to identify new biological targets for drug development more quickly and accurately.
2. Predictive Modeling: Machine learning models can predict how different compounds will interact with targets, helping to prioritize candidates for further testing and reducing the number of compounds that need to be synthesized and tested in the lab.
3. Optimized Clinical Trials: AI can assist in designing more efficient clinical trials by identifying suitable patient populations, predicting outcomes, and monitoring trial data in real-time, which can reduce costs and timelines.
4. Personalized Medicine: AI can analyze patient data to tailor drug therapies to individual patients, improving efficacy and reducing adverse effects, thereby advancing the concept of precision medicine.
5. Drug Repurposing: AI can facilitate the identification of new uses for existing drugs by analyzing existing data, which can lead to faster development timelines and lower costs compared to developing new compounds from scratch.
6. Automated Drug Discovery: Robotics combined with AI can automate high-throughput screening processes, speeding up the discovery of viable drug candidates.
7. Natural Language Processing (NLP): AI-driven NLP can help sift through the vast literature and clinical trial data to extract insights and trends that inform drug development strategies.
8. Regulatory Support: AI can help streamline regulatory processes by providing robust data analysis and insights that support faster approval timelines, potentially improving the time-to-market for new therapies.
9. Continuous Learning: AI systems can continuously learn from new data, enabling them to refine their predictions and models over time, which enhances their effectiveness in drug discovery.
10. Collaborative Platforms: The future may see increased collaboration among academia, industry, and technology companies, leveraging AI to pool resources and expertise for drug discovery.

Overall, as AI technology continues to evolve, its integration into drug discovery and development processes is likely to lead to more efficient, innovative, and effective therapeutic solutions.

Ethical Issues & Regulations

The integration of AI in drug development brings several ethical issues and regulatory challenges. Here are some key considerations:

Ethical Issues

1. Data Privacy and Security: AI systems often require access to vast amounts of patient data. Ensuring the privacy and security of this data is paramount to prevent breaches and misuse.
2. Bias and Fairness: AI models can perpetuate or even exacerbate existing biases in healthcare data. If not carefully managed, this can lead to inequitable treatment outcomes across different demographic groups.
3. Transparency and Accountability: The "black box" nature of some AI algorithms can make it difficult to understand how decisions are made. This lack of transparency raises questions about accountability, especially if an AI system leads to adverse patient outcomes.
4. Informed Consent: Using AI in drug development may complicate the informed consent process, particularly regarding how patient data is used and the potential risks involved in AI-driven therapies.
5. Intellectual Property: As AI generates new drug candidates and insights, questions arise about who owns these discoveries and how to protect intellectual property in an increasingly automated process.
6. Human Oversight: The role of human oversight in decision-making is crucial. There is a risk that over-reliance on AI may diminish critical thinking and expert judgment in drug development processes.

Regulatory Challenges

1. Lack of Clear Guidelines: Regulatory frameworks for AI in drug development are still evolving. There is a need for clear guidelines on how AI models should be validated and used in the drug approval process.
2. Regulatory Approval Processes: Traditional drug approval processes may not adequately accommodate the rapid iteration and learning capabilities of AI. Regulatory bodies need to adapt to ensure that AI-driven drugs are safe and effective.
3. Validation of AI Models: Ensuring that AI models are rigorously validated is critical. Regulatory agencies will need to establish standards for evaluating the performance and reliability of these models before they can be used in drug development.
4. Interdisciplinary Collaboration: Effective regulation will require collaboration between AI experts, healthcare professionals, and regulatory bodies to understand the implications of AI in drug development fully.
5. Post-Market Surveillance: Ongoing monitoring of AI-driven therapies will be essential to detect unforeseen issues and ensure long-term safety and efficacy.

Case Studies of Successful AI-Aided Drug Discovery Efforts

The potential of AI in the context of drug discovery has been demonstrated in several case studies. For example, the successful use of AI to identify novel compounds for the treatment of cancer has recently been reported by Gupta, R., et al. [24]. These authors trained a DL algorithm on a large dataset of known cancer-related compounds and their corresponding biological activity. As an output, novel compounds with high potential for future cancer treatment were obtained, demonstrating the ability of this method to discover new therapeutic candidates. The use of ML to identify small-molecule inhibitors of the protein MEK [25] has recently been described. MEK is also a possible target for the treatment of cancer, but the development of effective inhibitors has been challenging. The ML algorithm was able to identify novel inhibitors for this protein. Another example is the identification of novel inhibitors of beta-secretase (BACE1), a protein involved in the development of Alzheimer’s disease [26] by using an ML algorithm. AI has also been successfully applied in the discovery of new antibiotics [27]. A pioneering ML approach has identified powerful types of antibiotic from a pool of more than 100 million molecules, including one that works against a wide range of bacteria, such as tuberculosis and untreatable bacterial strains [28]. The use of AI in the discovery of drugs to combat COVID-19 has been a promising area of research during the last two years. ML algorithms have been used to analyze large datasets of potential compounds and identify those with the most potential for treating the virus. In some cases, these AI-powered approaches have been able to identify promising drug candidates in a fraction of the time that it would take when using traditional methods [29,30,31,32,33,34].

CONCLUSION

However, the successful application of AI in drug discovery is dependent on the availability of high-quality data, the addressing of ethical concerns, and the recognition of the limitations of AI-based approaches. The advancement of AI, along with its remarkable tools, continuously aims to reduce challenges faced by pharmaceutical companies, impacting the drug development process along with the overall lifecycle of the product, which could explain the increase in the number of start-ups in this sector AI can also make major contributions to the further incorporation of the developed drug in its correct dosage form as well as its optimization, in addition to aiding quick decision-making, leading to faster manufacturing of better-quality products along with assurance of batch-to-batch consistency. AI can also contribute to establishing the safety and efficacy of the product in clinical trials, as well as ensuring proper positioning and costing in the market through comprehensive market analysis and prediction.

REFERENCES

1. Gupta D., Suryarao S. , “DRUG DEVELOPMENT: STAGES OF DRUG DISCOVERY AND DEVELOPMENT PROCESS”, JETIR August 2022, Volume 9, Issue 8.
2. Hughes JP, Rees S, Kalindjian SB, Philpott KL. 2011,Principles of early drug discovery. Br J Pharmacol,162: 1239-1249.
3. Pharmaceutical Research and Manufacturers Association (PREMA) Introduction to drug development process. 30th PREMA Anniversary.
4. Catrin H. and Tudor I., “Artificial Intelligence for Drug Discovery: Are We There Yet?” Annual Review of Pharmacology and Toxicology, September 22, 2023: 528
5. Debleena Paulz, Gaurav Sanap, “Artificial intelligence in drug discovery and development, elesvier, vol:26 January,2021.umber 1  January 2021. I
6. Mohamad A.and Kaul. N, What are the applications of artificial intelligence in drug discovery & development? By prescouter, August,2018 p.g.8
7. Niazi SK (2023) The Coming of Age of AI/ML in Drug Discovery, Development, Clinical Testing, and Manufacturing: The FDA Perspectives. Drug Des Devel Ther 17: 2691-2725.
8. Paul D, Gaurav Sanap, Snehal Shenoy, Dnyaneshwar Kalyane, Kiran Kalia, et al. (2021) Artificial intelligence in drug discovery and development. Drug Discov Today 26(1): 80-93.
9. Blanco González A, Alfonso Cabezón, Alejandro Seco González, Daniel Conde Torres, Paula Antelo Riveiro, et al. (2023) The Role of AI in Drug Discovery: Challenges, Opportunities, and Strategies. Pharmaceuticals (Basel) 16(6): 891.
10. Askr H, Enas Elgeldawi, Heba Aboul Ella, Yaseen AM Elshaier, Mamdouh M Gomaa, et al. (2023) Deep learning in drug discovery: an integrative review and future challenges. Artif Intell Rev 56(7): 5975-6037.
11. Gupta R, Devesh Srivastava, Mehar Sahu, Swati Tiwari, Rashmi K et al. (2021) Artificial intelligence to deep learning: machine intelligence approach for drug discovery. Mol Divers 25(3): 1315-1360.
12. https://www.kandasoft.com/ai-and-its-impact-on-drug-development-benefits-challenges-and-use-cases/
13. https://www.altexsoft.com/blog/ai-drug-discovery-repurposing/
14. Zhu, H. (2020) Big data and artificial intelligence modeling for drug discovery. Annu. Rev. Pharmacol. Toxicol. 60, 573–589
15. Ciallella, H.L. and Zhu, H. (2019) Advancing computational toxicology in the big data era by artificial intelligence: data-driven and mechanism-driven modeling for chemical toxicity. Chem. Res. Toxicol. 32, 536–547
16. Chan, H.S. et al. (2019) Advancing drug discovery via artificial intelligence. Trends Pharmacol. Sci. 40 (8), 592–604
17. Brown, N. (2015) Silico Medicinal Chemistry: Computational Methods to Support Drug Design. Royal Society of Chemistry
18. Pereira, J.C. et al. (2016) Boosting docking-based virtual screening with deep learning. J. Chem. Inf. Model. 56, 2495–2506.
19. Vamathevan, J.; Clark, D.; Czodrowski, P.; Dunham, I.; Ferran, E.; Lee, G.; Li, B.; Madabhushi, A.; Shah, P.; Spitzer, M. Applications of machine learning in drug discovery and development. Nat. Rev. Drug Discov. 2019, 18, 463–477. [CrossRef] [PubMed
20. Gómez-Bombarelli, R.;Wei, J.N.; Duvenaud,D.;Hernández-Lobato, J.M.; Sánchez-Lengeling, B.; Sheberla,D.;Aguilera-Iparraguirre, J.; Hirzel, T.D.; Adams, R.P.; Aspuru-Guzik, A. Automatic chemical design using a data-driven continuous representation of molecules. ACS Cent. Sci. 2018, 4, 268–276. [CrossRef]
21. Tropsha, A. Best Practices for QSAR Model Development, Validation, and Exploitation. Mol Inf. 2010, 29, 476–488. [CrossRef] [PubMed]
22. Schneider, G. Automating drug discovery. Nat. Rev. Drug Discov. 2018, 17, 97–113. [CrossRef] [PubMed]
23. Rushikesh Dhudum 1, Ankit Ganeshpurkar 2 and Atmaram Pawar 3,* Revolutionizing Drug Discovery: A Comprehensive Review of AI Applications, 13 February 2024, 165.
24. Gupta R., Srivastava D., Sahu M., Tiwari S., Ambasta R.K., Kumar P. Artificial intelligence to deep learning: Machine intelligence approach for drug discovery. Mol. Divers. 2021;25:1315–1360. doi: 10.1007/s11030-021-10217-3. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
25. Zhu J., Wang J., Wang X., Gao M., Guo B., Gao M., Liu J., Yu Y., Wang L., Kong W., et al. Prediction of drug efficacy from transcriptional profiles with deep learning. Nat. Biotechnol. 2021;39:1444–1452. doi: 10.1038/s41587-021-00946-z. [PubMed] [CrossRef] [Google Scholar]
26. Dhamodharan G., Mohan C.G. Machine learning models for predicting the activity of AChE and BACE1 dual inhibitors for the treatment of Alzheimer’s disease. Mol. Divers. 2022;26:1501–1517. doi: 10.1007/s11030-021-10282-8. [PubMed] [CrossRef] [Google Scholar]
27. Melo M.C.R., Maasch J.R.M.A., de la Fuente-Nunez C. Accelerating antibiotic discovery through artificial intelligence. Commun. Biol. 2021;4:1050. doi: 10.1038/s42003-021-02586-0. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
28. Marchant J. Powerful antibiotics discovered using AI. Nature. 2020. Online ahead of print . [PubMed] [CrossRef]
29. Lv H., Shi L., Berkenpas J.W., Dao F.Y., Zulfiqar H., Ding H., Zhang Y., Yang L., Cao R. Application of artificial intelligence and machine learning for COVID-19 drug discovery and vaccine design. Brief. Bioinform. 2021;22:bbab320. doi: 10.1093/bib/bbab320. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
30. Monteleone S., Kellici T.F., Southey M., Bodkin M.J., Heifetz A. Methods in Molecular Biology. Volume 2390. Humana Press Inc.; Totowa, NJ, USA: 2022. Fighting COVID-19 with Artificial Intelligence; pp. 103–112. [PubMed] [Google Scholar]
31. Zhou Y., Wang F., Tang J., Nussinov R., Cheng F. Artificial intelligence in COVID-19 drug repurposing. Lancet Digit. Health. 2020;2:e667–e676. doi: 10.1016/S2589-7500(20)30192-8. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
32. Verma N., Qu X., Trozzi F., Elsaied M., Karki N., Tao Y., Zoltowski B., Larson E.C., Kraka E. Predicting potential Sars-Cov-2 drugs-in depth drug database screening using deep neural network framework ssnet, classical virtual screening and docking. Int. J. Mol. Sci. 2021;22:1392. doi: 10.3390/ijms22031392. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
33. Bung N., Krishnan S.R., Bulusu G., Roy A. De novo design of new chemical entities for SARS-CoV-2 using artificial intelligence. Future Med. Chem. 2021;13:575–585. doi: 10.4155/fmc-2020-0262. [PMC free article] [PubMed] [CrossRef] [Google Scholar]