• info@icbcongress.com
• English
• Persian

• Login

Create Free Account

INTERNATIONAL CONGRESS OF

PharmacoGenetics

• Pharmacogenomics and Artificial Intelligence: current application and the future integration of AI in pharmacogenomic aspect

• Saina Adiban Afkham,^1,* Sadaf Hasrati,²
  1. Systems Artificial Intelligence Network (SAIN), Universal Scientific Education and Research Network (USERN), Tehran, Iran
  2. Department of genetics, Modern sciences and technologies faculty, Islamic Azad University, Tehran Medical Branch, Tehran, Iran
  
  
  
• Introduction: The integration of artificial intelligence (AI) in healthcare, particularly in cancer treatment, is transforming the landscape of precision medicine. AI programs are designed to assist clinicians with diagnosis, treatment decisions, and outcome predictions, thereby enhancing patient care through improved data management and knowledge application. A significant aspect of this integration is pharmacogenomics, which studies how genetic variations influence drug responses, allowing for tailored treatment strategies. Pharmacogenomics plays a crucial role in identifying genetic markers that determine drug efficacy and toxicity. By incorporating genomic data into healthcare systems, AI can help predict patient responses to medications, facilitating early diagnosis and enabling pharmaceutical companies to target treatments more effectively. The identification of genomic markers, such as polymorphisms or gene expression patterns, is essential for predicting drug responses, and various methodologies, including Genome-Wide Association Studies, are employed to uncover these markers. Deep learning (DL) techniques are increasingly utilized in clinical oncology to analyze genomic, transcriptomic, and histopathological data. These tools serve as decision-support systems for healthcare professionals, enhancing their ability to diagnose and manage cancer patients. The goal is not to replace human expertise but to augment it, providing researchers and clinicians with powerful tools to improve patient outcomes. Recent research highlights the development of deep neural network models that can extract critical features from genetic mutations and gene expression, bridging the gap between preclinical findings and clinical applications. For instance, studies have shown that AI can effectively translate pharmacogenomic insights from cell line models to predict drug responses in actual tumors. Moreover, AI's potential extends to drug discovery and development, where it can analyze vast datasets to identify promising drug candidates and optimize existing therapies. This capability not only accelerates the discovery of new treatments but also enhances the precision of existing ones, ultimately leading to better patient outcomes. In summary, the fusion of AI and pharmacogenomics is paving the way for a new era in cancer treatment, characterized by personalized medicine that considers individual genetic profiles to optimize therapeutic strategies and improve patient care.
• Methods: A comprehensive database of clinical, in vivo, and in vitro studies published in English between 2004 and 2023 was identified through PubMed. Search terms included pharmacogenomics, pharmacogenetics, artificial intelligence, machine learning and deep learning.
• Results: A PubMed search using specific terms yielded 515 results. Following a thorough review process, 35 publications were selected for inclusion in our research. The inclusion criteria required that the publications be relevant to the title and abstract. The exclusion process involved evaluating each publication based on predefined criteria and removing duplicates. Additionally, the inability to access the full text of some publications resulted in a final selection of 35 publications for our study.
• Conclusion: The digitalization of patient records enables centralized data storage and offers opportunities for gathering additional information. This also adds complexity to education, prediction, and diagnosis of medical conditions. To manage this expansion, more advanced AI technologies such as Machine Learning and Deep Learning are necessary to help healthcare professionals extract valuable details from the data. Consequently, AI is anticipated to have a significant impact on healthcare information systems, potentially providing assistance or partially substituting for medical experts in the future. Integrating genomic data into a healthcare system based on knowledge is a method to fully leverage that information into improving patient care. As a result of multi-omics analysis becoming popular, multimodal learning techniques are anticipated to be more common in cancer diagnosis. Yet, the difficulties associated with obtaining multi-omics data from patient samples in a clinical setting, rather than from samples stored for research purposes, could hinder the practical application of these methods in clinical environments. In conclusion, the combination of pharmacogenomics and AI represents a promising frontier in cancer treatment. Continued advancements in this field will likely lead to more personalized, effective, and safer therapeutic options for cancer patients, ultimately contributing to better health outcomes and quality of life.
• Keywords: pharmacogenomics, pharmacogenetics, artificial intelligence, machine learning, deep learning