Glioblastoma (GBM) is a complex and confounding disease that works quickly and whose prognosis has not seen much improvement in 30 years; patients with aggressive glioblastoma, and who respond to treatment, have a median survival rate of about 14 months. With the technologies and analyses that are available today through genomics – including IBM’s Watson – there is new promise. The GBM study aims to develop clinically actionable applications that are scalable and can be offered to current GBM patients.Pphysicians and scientists are involved with the Glioblastoma Outcomes Trial, from neurosurgeons to biostatisticians to bioethicists, across nine different institutions and NYGC: The Rockefeller University, Memorial Sloan-Kettering, Weill Cornell, North Shore-LIJ, NYU Langone, Columbia, Albert Einstein, Montefiore, and Roswell Park.

We are in the middle of building a phenotype registry. Our approach is to use REDCap to collect data and build a semantic layer (ontology) on top of its unstructured data. Our goal is to use this project as a study case to build a generic workflow methods to integrate REDCap or other database information into the ontology for phenotype-genotype analysis.

Design Diagram (click to enlarge):

Team members: Mayu Okawa Frank (NYGC & Darnell Lab),Cameron Coffran (Medical Informatics), Evan Noch (Weill Cornell), Siddhartha Mitra (Research Bioinformatics), Yupu Liang (Research Bioinformatics)