Simeon Petrinin

How Clinical Model Composition Influences the Prognostic Utility of Transcriptomic Data in Cancer Survival Prediction

Started Mar. 6, 2026

Abstract or project description

Simeon will be evaluating whether adding additional omics data improves survival predictions in a machine learning model over conventional clinical metadata. He will start with evaluating this in lung cancer and will expand to additional cancer types if type allows. To accomplish this, he will focus on publicly available clinical, transcriptomic, pathology, and maybe genomic data from TCGA. This research will be important and interesting in evaluating the value of addition the additional complexity and resource-intensive collection of these types of data for patients.