Dana Silverbush

Dana started as a computer scientist with a B.S., M.S., and PhD from the Computer Science School at Tel Aviv University, focusing on deciphering abnormal signaling in cancer using AI, ML, and network approaches. Wanting to bridge the gap between research and the clinic, she joined as a shared postdoc between the Broad Institute of MIT and Harvard and Massachusetts General Hospital. Her postdoctoral studies involved developing new approaches for multi-omic single-cell techniques and studying the interlinks between genetics, DNA methylation, and transcription.

Translating these findings into clinical applications, Dana developed an ML approach that computationally predicts the tumor composition of a GBM, including both intratumoral malignant heterogeneity and microenvironment composition. This enhances diagnosis and prognosis capabilities. An exciting application of the method revealed a subset of patients who would benefit from a specific chemotherapy regimen versus those who would not. In another study, it highlighted a subgroup of patients who can benefit from maximal resection of their tumors versus those who would face an increased risk of rapid recurrence and epileptic events – all based on analyzing the patient through the lenses of their tumor heterogeneity and plasticity.