• info@icbcongress.com
• English
• Persian

• Login

Create Free Account

INTERNATIONAL CONGRESS OF

CancerGenetics

• Artificial Intelligence-Based Epigenetic Models for Predicting Hereditary Cancer Susceptibility

• Negar khaki,^1,* Kiana hajavi ,²
  1. Medical university and Azad university
  2. Azad university
  
  
  
• Introduction: Hereditary cancers such as breast, ovarian, and colorectal cancers are often driven by germline mutations; however, recent studies highlight the role of epigenetic alterations—particularly DNA methylation—in inherited cancer susceptibility. Epigenetics refers to heritable changes in gene expression that do not involve changes to the DNA sequence. The integration of artificial intelligence (AI) into epigenomic analysis enables the identification of subtle methylation patterns that may help predict cancer risk more accurately. This study explores how AI-based models can analyze epigenetic signatures to stratify hereditary cancer risk.
• Methods: We reviewed recent advances in AI-powered epigenomic tools, focusing on machine learning (ML) and deep learning (DL) approaches such as support vector machines (SVMs), random forests, convolutional neural networks (CNNs), and graph neural networks (GNNs). These models were evaluated for their ability to process DNA methylation data and classify cancer risk. We also examined multi-cancer early detection (MCED) tests and multi-omics integration strategies from recent peer-reviewed studies.
• Results: AI-based epigenetic models demonstrated high accuracy in identifying cancer-specific methylation patterns, with deep learning models like CNNs and EMethylNET showing superior performance. Multi-omics approaches that integrate genomic and epigenomic data further improved risk prediction. Tools like GRAIL’s Galleri test and UCLA’s AI-driven epifactor model outperformed traditional diagnostic criteria by offering early-stage detection and outcome prediction across diverse cancer types.
• Conclusion: Artificial intelligence has shown strong potential in transforming cancer risk assessment by decoding complex epigenetic patterns. Despite challenges such as data heterogeneity and model interpretability, the continued evolution of explainable AI (XAI) and integration of multi-omics data promises to enhance the precision and accessibility of hereditary cancer prediction. AI-driven epigenomic models may soon become essential tools in personalized medicine and targeted surveillance.
• Keywords: Epigenetics, Artificial Intelligence, DNA Methylation, Hereditary Cancer, Deep Learning, Multi-omics