Graph4Med

Background

Medical databases normally contain large amounts of data in a variety of forms. With the rapid advancement of biomedical technologies, new forms of data from methods such as Next Generation Sequencing (NGS) or chromosome microarray (array CGH) are constantly being generated. Although they grant significant insights into diagnosis and treatment implementing them into current medical databases is challenging since these are most often based on a relational database schema and cannot be used to easily extract information for cohort analysis and visualization. As a consequeunce, valuable information regarding cohort distribution or patient similarity may be missed.

Description

We present Graph4Med, a web application that relies on a Neo4J graph database obtained by transforming a traditional relational database. Graph4Med provides a straightforward visualization and analysis of a selected patient cohort. Our original use case was a database of pediatric ALL (Acute Lymphoblastic Leukemia). We developed a suitable graph data schema to convert the relational data into a graph data structure using Neomodel and store it in a Neo4j database. Then, we used Neodash to build a dashboard for querying and displaying patients’ cohort analysis.

Conclusion

Our tool is capable of:

1. Quickly displaying the overview of patients’ cohort information such as distributions of gender, age, mutations (fusions), diagnosis
2. Providing mutation (fusion) based similarity search and displaying them in a maneuverable graph
3. Generating an interactive graph of any selected patient and facilitating the identification of interesting patterns among patients