Embargo until
Journal Title
Journal ISSN
Volume Title
Johns Hopkins University
Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest forms of cancer in the United States and is often diagnosed in advanced stages with poor prognosis. A new workflow called CODA that uses machine learning to reconstruct pancreas pathology, from precursor lesions to PDAC, has been established to study PDAC in humans in three-dimensions. Although the genetic mutations that drive PDAC are known, there exists little information regarding 3D spatial distribution of these mutations. Once defined in 3D, these mutations would need to be visualized in a clear and organized way. The application of genetic sequencing to 3D-constructed precursor lesions in the human pancreas afforded a novel opportunity to develop tools to visualize complex genetic changes in three dimensions. Each lesion was subdivided for deeper resolution of lesion heterogeneity. The visualization developed took a 3D scatter plot approach. Genetic mutations were represented by mapped objected spaced equally throughout the precursor lesions. Each genetic mutation was assigned a color. Object size was used to represent prevalence of each genetic mutation in 4 distinct 3D precursor lesions in each gene sequencing region. Four visualization outputs were created, including still images, turntable videos, an interactive platform, and a promotional image. The interactive platform includes a 3D interactive model that a user can rotate and scale, togglable genetic mutation representations, and a switch between “prevalence” and “no prevalence” modes. Modeling was done using 4D® and ZBrush®. Unity was used for lighting, materials, and creation of the 3D interactive platform. This thesis project experimented with ways in which data commonly visualized in a 2D manner could be visualized in a 3D space. The visualization represents a first step in understanding tumorigenesis in three dimensions and its contributing factors as related to tumor microenvironments in human.
pancreatic cancer, pancreatic ductal adenocarcinoma, PDAC, pancreatic intraepithelial neoplasia, PanIN, genetic mutation, KRAS, GNAS, RET, TP53, 3D, data visualization