• info@icbcongress.com
• English
• Persian

• Login

Create Free Account

INTERNATIONAL CONGRESS OF

PharmacoGenetics

• AI-Enhanced Pharmacogenomics: Toward Clinically Actionable and Personalized Drug Therapy

• Mehrsa Karim,¹ Mina Shirmohammadpour,² Bahman Mirzaei,^3,*
  1. Department of Microbiology and Virology, Faculty of Medicine, Zanjan University of Medical Sciences, Zanjan, Iran
  2. Department of Microbiology and Virology, Faculty of Medicine, Zanjan University of Medical Sciences, Zanjan, Iran
  3. Department of Microbiology, Faculty of Medicine, Urmia University of Medical Sciences, Urmia, West Azerbaijan, Iran
  
  
  
• Introduction: Inter-individual variability in drug response remains a major barrier to effective pharmacotherapy. While pharmacogenetics offers a framework for precision dosing, conventional statistical models are limited in handling complex genomic interactions. This study investigates the application of artificial intelligence (AI) to integrate genomic and multi-omics datasets for predicting drug response and adverse reactions.
• Methods: Genomic, transcriptomic, and clinical pharmacogenetic data collected between 2018 and 2024 were analyzed using deep learning architectures, including variational autoencoders and graph neural networks. Models were trained on publicly available datasets (e.g., PharmGKB, TCGA) and validated against independent clinical cohorts. Performance metrics included predictive accuracy, sensitivity, and interpretability.
• Results: AI-based models outperformed conventional regression methods, improving predictive accuracy by 12–18% across multiple drug–gene pairs. Multi-omics integration enhanced biomarker identification, particularly in oncology, where AI-driven models successfully stratified responders to tyrosine kinase inhibitors and immunotherapies. Generative AI tools enriched with CPIC guidelines demonstrated ~90% concordance with expert therapeutic recommendations. However, challenges were noted in harmonizing heterogeneous data, ensuring model interpretability, and mitigating algorithmic bias.
• Conclusion: Our findings highlight the potential of AI to advance pharmacogenetics by delivering clinically actionable insights and personalized therapeutic strategies. Future work should emphasize explainable AI, federated learning, and real-world clinical trials to ensure safe and equitable implementation of AI-driven pharmacogenomics in healthcare.
• Keywords: Pharmacogenetics, artificial intelligence, genomics ,multi-omics, precision medicine