Finding the Match: The Data & Science Behind Stem Cell Donor Selection for Blood Cancer Patients
Industrial Problems Seminar
Abeer Madbouly
NMDP
Abstract
Blood cancers (Such as Leukemias, Lymphomas, and Myeloma) or hematologic genetic disorders (such as Sickle Cell Disease and Thalassemia) are often difficult to treat using pharmaceutical drugs or other treatment regimens like radiation or chemotherapy and often the only cure is a stem cell transplant. The donor of stem cells could be related or unrelated to the patient. The donation process requires a tissue match of the donor and the patient for successful transplant outcomes. The tissue match is complex, involves multiple genes and requires the application of complex statistical, bioinformatics, GIS and machine-learning models to handle data gaps and ambiguities, efficiently manipulate large repositories of data and deliver quick and accurate solutions to transplant centers serving patients worldwide. In this presentation, the students will learn about different algorithms developed by the Bioinformatics, Data Analytics and Statistics teams at NMDP, the US National Stem-Cell Donor Registry, to streamline the process of donor selection, donor/patient matching, predict donor availability and even plan for strategic donor recruitment.