Integrating artificial intelligence (AI) into drug discovery revolutionizes pharmaceutical research by significantly accelerating the identification, optimization, and development of novel therapeutics. Conventional drug discovery methods, known for high costs, lengthy timelines, and low success rates, are increasingly being augmented by AI-based technologies, including machine learning (ML), deep learning (DL), and reinforcement learning (RL). These advanced computational approaches enhance key processes, such as target identification, virtual screening, de novo drug design, toxicity prediction, and the optimization of pharmacokinetic and pharmacodynamic profiles, dramatically increasing overall efficiency. AI-driven primary and secondary screening methods improve cell classification, compound prioritization, and drug-target interaction predictions, substantially shortening the progression from preclinical phases to clinical trials. Additionally, AI enables retrosynthesis prediction and reaction yield modeling, optimizing chemical synthesis pathways and reducing the need for resource-intensive experimental procedures. AI's integration into clinical trials has notably improved patient stratification, biomarker discovery, and adaptive trial designs, ultimately delivering more precise and economically feasible therapeutic interventions. Furthermore, AI supports polypharmacological approaches through multitarget drug discovery, drug repurposing (finding new uses for existing drugs), and adverse effect prediction, thereby advancing personalized medicine. Despite these transformative advantages, it's important to note that AI in drug discovery also has limitations, such as ensuring data quality, improving model interpretability, gaining regulatory acceptance, and addressing ethical concerns. This review comprehensively explores the impact of AI throughout the drug discovery pipeline, emphasizing its critical role in expediting the development of life-saving medications and outlining future directions for continued pharmaceutical innovation driven by AI.

The conventional drug discovery pipeline is labour-intensive, time-consuming, and costly, involving target identification, hit discovery, lead optimization, and extensive preclinical and clinical evaluation. To overcome these limitations, artificial intelligence (AI) has emerged as a transformative tool in drug discovery, gaining widespread adoption in the pharmaceutical industry during the 2010s due to advances in computing power, data availability, and deep learning. AI-based approaches, including molecular property prediction, protein structure modelling, natural language processing, and ADME/Tox prediction, have enhanced efficiency, reduced costs, and improved decision-making across multiple stages of drug development. Several AI-guided molecules have progressed into clinical trials, with encouraging early-phase success rates, highlighting the potential of AI to accelerate innovation. However, despite more than a decade of intensive research, no AI-only originated drug has yet achieved full regulatory approval, reflecting persistent challenges consistent with Eroom's law. Key limitations include poor data quality and accessibility, lack of model interpretability, gaps between computational predictions and chemical feasibility, and the inherent complexity of biological systems that limit translational success. Furthermore, AI-driven hypothesis generation does not replace the need for scientific reasoning and experimental validation. Overall, while AI has significantly accelerated early drug discovery stages, it remains a supportive tool rather than a standalone solution, underscoring the continued need for human expertise and experimental research.

Quantitative prediction of human pharmacokinetic drug-drug interactions and drug clearance using humanized liver chimeric mice: a review.

Why are there no clinically-approved drugs targeting disordered proteins?

Natural products are vital sources of medicines, and their analogues hold great promise as novel therapeutics to combat disease more effectively, including circumventing antibiotic resistance. Engineering biosynthetic pathways in microbial hosts enables efficient, cost-effective production of non-natural analogues - compounds structurally related to natural products but not necessarily derived through direct chemical modification. In this review, we highlight recent work on the in vivo production of novel natural product analogues, with an emphasis on combinatorial and precursor-directed biosynthesis, enzyme engineering, and retrobiosynthesis. We anticipate that further developments in artificial intelligence, particularly the use of machine learning models to understand enzymatic transformations and predict novel reactions, will significantly accelerate this field and drive forward its importance in drug discovery and related research.

PROTACs as novel therapeutics against Mycobacterium tuberculosis: Current progress and future directions.

Efficient drug-target affinity prediction via interaction features and parallel CNN-BiLSTM with attention.

SSHF-DTI: Leveraging structural similarity and hierarchical features through a fusion network for drug-target interaction prediction.

Targeting Mycobacterium tuberculosis GAPDH elicits potent bactericidal responses by dysregulating enzyme activity, redox dynamics and iron acquisition.

Millimeter-scale, high-density three-dimensional constructs recapitulate hot and cold tumor microenvironment.

Recent advancements in liquid chromatography column technologies: Manufacturing, connecting and parallel using.

Proline, a privileged fragment in drug design: advances and future perspectives.

Recent advances in chemistry and bioactivity of decalin-containing natural products (2014-2025).

UDP-Glucuronosyl transferase mediated drug-drug interactions: An Industry perspective on recommended in vitro studies.

Nicotinic acetylcholine receptors-targeting drug discovery.

Lupenone regulates LOXL2-mediated PANoptosis signaling through E3 ubiquitin ligases RNF168 to improve radiation-induced lung injury.

Discovery of a novel compound against BmNPV using virtual screening based on the protein structure of the viral whole genome.

The TP53 gene encodes the tumor suppressor protein 53, which is critical for maintaining genomic stability and preventing tumorigenesis. Mutations in TP53, particularly missense mutations, have a substantial impact on cancer progression because they give gain-of-function features that enhance proliferation, metastasis, and treatment resistance. This review examines the mechanisms underlying p53 mutations, including their interactions with critical regulatory circuits, and assesses novel medication and prodrug options targeting TP53 mutations in various malignancies. Small-molecule correctors, statins, Hsp90 inhibitors, and new drugs like Eprenetapopt and COTI-2 are among the therapeutic options proposed. The mechanisms of action and potential efficacy in treating leukemia, lung, breast, and ovarian malignancies are investigated. Emerging techniques for restoring wild-type p53 functionality or degrading mutant p53 demonstrate the therapeutic potential of these approaches. Challenges such as medication resistance, side effects, and the chemical complexity of p53 pathways are also addressed, emphasizing the importance of ongoing research. This review contributes to our understanding of TP53-targeted cancer medicines, offering hope for more innovative treatments with improved outcomes.

From single-target to multi-target drugs: The significance of formula-derived nanoparticle drug discovery (FDD) as a novel paradigm for complex disease therapy

AI-powered literature mining reveals the therapeutic significance of GLP-1 receptor: Simulation of natural agonist candidates based on molecular dynamics.