Unveiling Biomedical Insights: A Semantic Pipeline from Document Entities to Knowledge Graph Analysis of Hallmarks of Cancer

This study aims to build a knowledge graph specifically customized to the characteristics or hallmarks of cancer. The aim is to converge on semantically and contextually valuable information, ensuring the insights generated are meaningful and relevant to cancer biology. Domain specialists play a critical role in leading this process by offering in-depth biomedical knowledge, particularly in identifying entities of interest within hallmark classes. We apply the Unified Medical Language System (UMLS) to standardize these things, assuring consistency across biomedical nomenclature. Finally, the knowledge graph is generated based on the expert-selected entities, focusing on familiar key entities highlighting critical links within the hallmark class characteristics. This approach provides a greater understanding of cancer hallmarks through structured and standardized knowledge representation.