Predicting Resistance to Bcl2 antagonists using an In-Silico Functional Assay

Shannon Cosgrove

Mentor: Dr. Matthew McCoy, Innovation Center for Biomedical Informatics, Department of Oncology and Department of Biostatistics, Bioinformatics and Biomathematics, Georgetown University

Date/Time: August 25, 2020, 4:00pm

Abstract: Chronic Leukocytic Leukemia and many other types of cancer are reliant on overexpression of the anti-apoptosis Bcl2 protein for survival. Bcl2 functions as a part of the Bcl2 family in maintaining apoptotic homeostasis. Antagonists of Bcl2, more specifically a class of drugs named BH3 mimetics due to their homology with the BH3 domain of the Bcl2 family, are common in the treatment of cancers reliant on Bcl2 mutations. Variants within the Bcl2 molecule, such as G101V, F104L, and D103Y have been seen to confer resistance in patients. In this study, in-silico simulations of molecular binding between BH3 mimetics and Bcl2 variants using AutoDock Vina and the Snp2Sim workflow were performed and compared to wild-type binding simulations. The majority of known resistant variants were accurately represented in the in-silico binding simulation with few exceptions which may indicate a more complex mechanism not yet incorporated into the simulation workflow. These results provide insight into the importance of in-silico functional assays as a means for predicting resistance in patients with Bcl2-reliant cancers.