CNSC-44. BIOLOGICAL PATTERN DISCOVERY IN GLIOBLASTOMA

Abstract Glioblastoma(GBM - IDH wildtype) is an adult glioma, showing abysmal prognosis. Sequencing technologies show us the existence of a neurodevelopmental frame-work, upon which heterogeneous molecular patterns are overlaid. The tumor's functional capacity is the net result of extrinsic+intrinsic molecular factors interacting with each other. These factors affect how cells communicate and organize themselves into complex functional architectures that support their evolution, survival and resistance. Understanding the phenotypic significance of a tumor's architecture, will allow us to understand these processes and compute complex patterns in GBM that are "biologically-relevant". First, we showed, as a proof-of-concept, that "biologically-relevant" signatures are imprinted within the spatial organization of cells using spatial pixel analysis. We used a phase-contrast image dataset of glioblastoma stem cells grown in culture, imaged 4-12hrly, over 12-16 days. We applied 29 hand-engineered pixel features per image, deriving spatial pixel signatures for each of our 17’601 phase-contrast images. Using different computational analytical methods and gene expression from matched bulk RNA datasets, we showed that spatial pixel patterning follows biologically relevant phenotypes. We found samples of images with high PC2 scores had higher mesenchymal/microglia scores, as compared to samples of images with low PC2 scores, which showed higher neurodevelopmental signatures. In addition, we found that mathematical algorithms describing entropy, homogeneity, contrast and complexity were clearly enriched in specific biological groups. Hence, our study showed that the organizational/architectural patterns of biological entities show preservation of fundamental organizational principles. Building upon the above data (which is currently a manuscript under preparation), we will apply these principles towards pattern discovery of GBM tissues at multiple architectural hierarchies (i.e subcellular, cellular, stroma) using different GBM models, imaging and spatial transcriptomics. If structure equates to function, understanding GBM architecture will have important applications in surgical tool developments and understanding relevant mechanistic patterns in GBM.