Artificial Intelligence in Drug Discovery and Delivery: Advancements and Applications

Artificial Intelligence (AI) has made significant progress in drug discovery and drug delivery and has become an active area of research. The use of AI in drug delivery has gained significant attention, with the development of new technologies and algorithms that enable more efficient drug delivery. The history of AI in drug discovery can be traced back to the 1960s, and since then, AI has been used in various stages of drug discovery, including target identification, lead optimization, and drug design. AI can aid in different stages of drug development, including drug discovery, formulation, and optimization. AlphaFold is an AI-powered deep learning system that uses neural networks to predict the structure of proteins to design drug molecules. The ultimate goal of using AI in pharmaceuticals is to minimize cost and time and achieve exceptional results without the need for laboratory presence. However, experimental validation is still necessary to verify the accuracy and reliability of the AI-generated results.

Artificial Intelligence (AI) as the simulation of the human intelligence process by computers [1] has made significant progress in various fields, including drug discovery [2] and drug delivery [3]. The use of AI in drug discovery and drug delivery has become an active area of research recently and its potential to accelerate the process of drug development and enable the discovery of new and more effective drugs is becoming increasingly recognized. In recent years, the use of AI in drug delivery has gained significant attention, with the development of new technologies and algorithms that enable more efficient drug delivery.

Drug delivery is an essential aspect of drug development [4]. The effectiveness of a drug depends not only on its pharmacological properties but also on its ability to reach the target site in the body [5]. Over the years, various drug delivery systems have been developed to improve drug efficacy and reduce side effects. However, the development of new drug delivery systems is a complex and time-consuming process. The use of AI in drug delivery has the potential to accelerate the development of new drug delivery systems and improve drug efficacy. In brief, Drug Delivery means delivering the right amount of medication to the right place at the right time [6].

The history of AI in drug discovery can be traced back to the 1960s when the first computer program, DENDRAL, was developed in Stanford university to assist in drug discovery [7,8]. Since then, AI has been used in various stages of drug discovery, including target identification [9], lead optimization [10], and drug design [11]. In the early days, AI was mainly used to predict the pharmacological properties of compounds. However, with the advancement of technology, AI is now being used to predict the 3D structure of proteins and design drugs (ie: MolAICal software) that can target specific proteins in the body [12].

The stages of using artificial intelligence in drug development

Artificial Intelligence (AI) has become an emerging tool in the field of drug discovery and formulation. AI can aid in different stages of drug development, including drug discovery, formulation, and optimization. In the initial stage of drug discovery, AI algorithms can be used to predict the potential drug targets and identify potential small molecule drug candidates. Medicinal chemistry has been using AI (ie: ChemDraw, Schrödinger Suite, Discovery Studio) for years to optimize drug structures and predict their pharmacological properties [13].

In the next stage of drug development, formulation, AI can aid in predicting the optimal drug delivery system. AI models can simulate drug release profiles and predict the stability of drug formulations under different environmental conditions. This can save time and costs in the experimental design phase of drug formulation. For example, software such as Design Expert can be used to optimize experimental designs [14].

Despite the benefits of using AI in drug discovery and formulation, experimental validation is still necessary to verify the accuracy and reliability of the AI-generated results. Therefore, AI and experimental designs should be used in tandem to accelerate the drug development process efficiently.

AlphaFold in drug discovery

One of the most promising recent developments in drug discovery developed by DeepMind is AlphaFold, an AI-powered deep learning system that uses neural networks to predict the structure of proteins to design drug molecules. By predicting the structure of proteins, it is possible to design drugs that target specific proteins in the body [15].

In a recent study titled "AlphaFold accelerates artificial intelligence-powered drug discovery: efficient discovery of a novel CDK20 small molecule inhibitor," the authors used AlphaFold to predict the structure of CDK20, a protein that is involved in the development of cancer. They then used this information to design a small molecule inhibitor that could target CDK20. The authors used a combination of virtual screening and experimental validation to identify the most promising inhibitor. They found that the inhibitor was highly effective in inhibiting the activity of CDK20 and showed promising results in preclinical studies [16]. In comparison with these advancements, the traditional AI drug delivery approaches are insufficient in tackling the challenges that arise from drug delivery due to their imprecision and inaccuracy in forecasting drug behavior inside the human body. However, novel AI-based approaches have the capacity to overcome these limitations by utilizing sophisticated algorithms capable of processing vast quantities of information to forecast drug behavior and optimize drug delivery. Moreover, AI-based approaches can provide tailored drug delivery by taking into account individual patient attributes, which can enhance drug effectiveness and minimize adverse reactions [3].

The ultimate goal of using artificial intelligence in pharmaceuticals is to minimize cost and time and achieve exceptional results without the need for laboratory presence. This ambitious goal entails leveraging AI algorithms to facilitate the generation of initial results for drug discovery, including the design and synthesis of novel drug molecules and formulation of drug candidates. AI can be employed to predict clinical outcomes and optimize drug efficacy, thus bringing life-saving medications to patients faster. According to the authors' belief, AI techniques, particularly machine learning, are not usually accompanied by fundamental concepts and laws to describe different phenomena and factors affecting performance improvement. For instance, in natural sciences such as physics and chemistry, there are fundamental concepts and laws to describe different phenomena, but in AI, complex and opaque models are often used, and for this reason, many people, including experts in this field, cannot fully understand why an AI model performs better in a particular case than in others and what factors have contributed to its improved performance.

1. Mak KK, Pichika MR. Artificial intelligence in drug development: present status and future prospects. Drug Discov Today. 2019 Mar;24(3):773-780. doi: 10.1016/j.drudis.2018.11.014. Epub 2018 Nov 22. PMID: 30472429.
2. Paul D, Sanap G, Shenoy S, Kalyane D, Kalia K, Tekade RKJDdt. Artificial intelligence in drug discovery and development. 2021;26(1):80.
3. He S, Leanse LG, Feng Y. Artificial intelligence and machine learning assisted drug delivery for effective treatment of infectious diseases. Adv Drug Deliv Rev. 2021 Nov;178:113922. doi: 10.1016/j.addr.2021.113922. Epub 2021 Aug 28. PMID: 34461198.
4. Langer R. New methods of drug delivery. Science. 1990 Sep 28;249(4976):1527-33. doi: 10.1126/science.2218494. PMID: 2218494.
5. Ding BS, Dziubla T, Shuvaev VV, Muro S, Muzykantov VR. Advanced drug delivery systems that target the vascular endothelium. Mol Interv. 2006 Apr;6(2):98-112. doi: 10.1124/mi.6.2.7. PMID: 16565472.
6. Wang H, Huang YJMidd. Combination therapy based on nano codelivery for overcoming cancer drug resistance. 2020;6:100024.
7. Schneider G. Automating drug discovery. Nat Rev Drug Discov. 2018 Feb;17(2):97-113. doi: 10.1038/nrd.2017.232. Epub 2017 Dec 15. PMID: 29242609.
8. Kavidopoulou A, Syrigos KN, Makrogkikas S, Dlamini Z, Hull R, Marima R, et al. AI and Big Data for Drug Discovery.  Trends of Artificial Intelligence and Big Data for E-Health: Springer; 2023;121-38.
9. Gupta R, Srivastava D, Sahu M, Tiwari S, Ambasta RK, Kumar PJMd. Artificial intelligence to deep learning: machine intelligence approach for drug discovery. 2021;25:1315-60.
10. Wang W, Ye Z, Gao H, Ouyang D. Computational pharmaceutics - A new paradigm of drug delivery. J Control Release. 2021 Oct 10;338:119-136. doi: 10.1016/j.jconrel.2021.08.030. Epub 2021 Aug 19. PMID: 34418520.
11. Zhang Y, Luo M, Wu P, Wu S, Lee TY, Bai C. Application of Computational Biology and Artificial Intelligence in Drug Design. Int J Mol Sci. 2022 Nov 5;23(21):13568. doi: 10.3390/ijms232113568. PMID: 36362355; PMCID: PMC9658956.
12. Bai Q, Tan S, Xu T, Liu H, Huang J, Yao XJBib. MolAICal: a soft tool for 3D drug design of protein targets by artificial intelligence and classical algorithm. 2021;22(3):bbaa161.
13. Zhou Y, Zhang Y, Lian X, Li F, Wang C, Zhu F, Qiu Y, Chen Y. Therapeutic target database update 2022: facilitating drug discovery with enriched comparative data of targeted agents. Nucleic Acids Res. 2022 Jan 7;50(D1):D1398-D1407. doi: 10.1093/nar/gkab953. PMID: 34718717; PMCID: PMC8728281.
14. Supare V, Wadher K, Umekar MJJoDD, Therapeutics. Experimental design: Approaches and applications in development of pharmaceutical drug delivery system. 2021;11(4-S):154-61.
15. Nussinov R, Zhang M, Liu Y, Jang HJTJoPCB. AlphaFold, artificial intelligence (AI), and allostery. 2022;126(34):6372-83.
16. Ren F, Ding X, Zheng M, Korzinkin M, Cai X, Zhu W, et al. AlphaFold accelerates artificial intelligence powered drug discovery: efficient discovery of a novel CDK20 small molecule inhibitor. 2023;14(6):1443-52.