Due to recent advancement in isolation and characterization of extracellular vesicles (EVs), their roles in different diseases including cancer is increasingly evident. Extracellular vesicles (EVs) hold promise as non-invasive biomarkers for disease diagnosis, particularly in cancer, due to their long distance cellular communication and regulation roles. With that, the field of EV is experiencing an influx of new data, including EV cargo proteome and RNA sequencing. This study proposes the development of a knowledge graph (KG) to integrate and analyze diverse EV data, including miRNA sequencing information obtained from publicly available resources. The KG captures complex relationships between EVs, microRNAs, disease conditions, and mRNA targets as nodes and edges and is designed to be stored in a graph database like Neo4j. By applying queries for retrieving all possible paths in the KG, the study aims to facilitate the discovery of novel relationships and insights into the functional roles of microRNAs within EVs. The KG has multiple benefits, that it supports the integration of multiple data sources for robust analysis, allows handling of heterogeneous nature of biomedical data, and provides easy ways of obtaining its projection for building machine learning pipelines. This study represents a step in the knowledge extraction process and the evaluation of graph model’s effectiveness in classification problems. We believe such study can pave the way towards more effective ways to incorporate graph models and network science in advancing biomedical research.