GENO-32AN ANATOMIC TRANSCRIPTIONAL ATLAS OF GLIOBLASTOMA

Abstract

The molecular and cellular landscape of glioblastoma is highly complex and its relationship to histologic features routinely used for diagnosis is unclear. To investigate this relationship, we generated an anatomic transcriptional atlas of human glioblastoma, adopting a highly-systematized, large-scale, histology-driven approach to the characterization of anatomic features and cancer stem cell populations. The atlas of 42 tumors consists of several data modalities, including 270 transcriptomes, ∼11,500 semi-annotated pathology images registered to ∼23,000 in situ hybridization gene expression images, ∼400 MRI scans, tumor-derived cell lines and xenografts, and supporting longitudinal clinical information. We show that gene expression varies considerably by anatomic feature and cancer stem cell population, exhibiting molecular signatures that are more highly conserved within tumors than between tumors, and reflecting the cell types and microenvironment of each feature and population. These freely-accessible online data resources of the Ivy Glioblastoma Atlas Project (Ivy GAP), one for the fully-annotated anatomic transcriptional atlas (http://glioblastoma.alleninstitute.org/), and one for the detailed clinical and genomic data (http://ivygap.org/), constitute a unique platform for exploring the anatomic and genetic basis of glioblastoma at the cellular and molecular levels. This project was supported by grants from the Ben and Catherine Ivy Foundation to the Allen Institute (PI, RBP) and Swedish Neuroscience Institute (PIs, GDF and NS).