Artificial Intelligence Based Drug Discovery, Virtual Surgery Software and Machine Learning in Pharmacological Targets Discovery

2026-01-05T15:48:19+00:00

The pharmaceutical industry is currently experiencing a transition crisis that has been called the “analog” to digital biology transition. The traditional drug development process is plagued by a high failure rate and a clinical success rate of about 10 - 12 per cent. Preclinical development expenses are usually more than $ 2.5 billion / new molecular entity. The technologies of Artificial Intelligence (AI) and Machine Learning (ML) have become groundbreaking, which can reverse this trend and bring a chance to reduce the development time-frames and expenditures to a significant margin. This is a review article on the critical and high-resolution integration of AI in the upstream of the drug discovery pipeline. The development of classical Quantitative Structure-Activity Relationship (QSAR) models via the state of the art deep Graph Neural Networks (GNNs) up to state of the art Transformer based models of pharmacological targets identification is theorized on and selectively analyzed. The revolutionary impact of Alpha-fold to the growth of the so-called druggable genome and applications of generative AI (such as Diffusion Models and Reinforcement Learning) to de novo molecular design are given special consideration. We also discuss the ease of the acceleration of virtual screening (VS) thanks to active learning schemes, which are performed with billions of chemical structures in chemical libraries. But lastly, we also openly address critical bottlenecks for widespread adoption (like bias of the data, model interpretability (Explainable AI) and the so-called hallucination of synthetically invalid structures) and outline the future of entirely autonomous and closed loop discovery laboratories.

Precision Pharmacotherapy: Pharmacogenomics, Biomarkers and Omics Technologies: Identification of Drugs and Doses in Complex Diseases in Patients

2026-03-17T18:46:20+00:00

The trends of a booming rate of therapeutic failure and drug reactions in the management of complex diseases, are causing the current clinical paradigm of the one-size-fits-all pharmacotherapy quickly going out of business. Precision pharmacotherapy is intended to replace this type of empiricism with a solution that uses medical therapy based on the biological profile of each patient. This literature review involves a thorough critical review of how pharmacogenomics, multi-omics integration, and artificial intelligence came together to redefine the selection of drugs and dose optimization. We review the move to genomic profiling in general rather than single gene testing and ascertain the clinical utility of germline variants in drug metabolizing enzymes and transporters. Besides, the recent contribution of transcriptomics and proteomics to the unravelling of the non-genetic factors of drug resistance are discussed, particularly in the area of oncology and immunology. The use of companion diagnostics is investigated as a critical process of a biomarker-based therapy, which requires that targeted agents will be withheld from a patient with necessary molecular pathology. In addition, we assess the potential of artificial intelligence algorithms to be used for the transformative interpretation of high-dimensional omics data to predict complex pharmacokinetic phenotypes. Lastly, the review discusses the high translational obstacles, in the fields of informatics, economics and ethics, to be overcome in order to allow the smooth assimilation of these technologies into the daily practice of clinical care. We conclude that multimodal precision-dosing framework is one of the essential requirements of the future sustainability of the healthcare system in the world.

International Journal of Drug Discovery and Medical Research

Artificial Intelligence Based Drug Discovery, Virtual Surgery Software and Machine Learning in Pharmacological Targets Discovery

Precision Pharmacotherapy: Pharmacogenomics, Biomarkers and Omics Technologies: Identification of Drugs and Doses in Complex Diseases in Patients