Potential for Drug Development Using Artificial Intelligence and Machine Learning

OBJECTIVES: The applications of artificial intelligence (AI) and machine learning (ML) programs in drug discovery have expanded dramatically. This study aims to systematically review and evaluate the efficacy of AI- and ML-based substances as potential candidates for new drugs.

METHODS: To identify pertinent studies, articles published up to May 2023 were retrieved from databases such as PubMed, CINAHL, Cochrane Library, and Web of Science, utilizing relevant keywords. Following the PICO (Population, Intervention, Comparator, Outcome) framework and adhering to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations, this study included articles evaluating the efficacy of AI- and ML-based substances in clinical trials.

RESULTS: The search strategy yielded 663 articles after removing duplicates, and eight studies met our inclusion criteria. Encompassing a variety of diseases and conditions, these studies examined the efficacy of peptides and molecules discovered through AI and ML. Notably, all peptides and molecules included in our analysis demonstrated efficacy for their respective targeted diseases. Among the eight articles, two studies evaluated the effectiveness of discovered peptides in treating type 2 diabetes by assessing changes in HbA1c levels. Another study validated a peptide's potential for treating microbial infections by observing colony forming units (CFUs) in bacterial samples. Additionally, two studies identified promising molecules for cancer treatment, assessing survival probability and tumor growth. One study validated a drug molecule for thrombocytopenia by measuring platelet regeneration. Lastly, two peptides targeting anti-aging and inflammation were evaluated for their effects on anti-aging and changes in TNF-α, glucose level, and serum LDL and HDL concentrations.

CONCLUSIONS: Although AI and ML in drug discovery are in early stages, the findings of this study suggest their potential to revolutionize the pharmaceutical industry and healthcare by accelerating drug development through the identification of unexplored molecules.