Abstract
We have developed a simple and generalizable in vivo method for brain tumor modeling, mosaic analysis by dual recombinase-mediated cassette exchange (MADR). MADR allows for stable labeling of mutant cells expressing transgenic elements from a precisely-defined chromosomal locus. We have demonstrated the power and versatility of MADR by creating novel glioma models with mixed, reporter-defined zygosity, or with “personalized” H3. 3-containing driver mutation signatures from pediatric glioma--each manipulation altering the spatiotemporal profile of resulting tumors. Further we have generated ependymoma models by employing patient-derived fusion driver mutations. Notably, each model displays divergent spatiotemporal tumor expansion profiles and cellular phenotypes. Now, we use single-cell RNA-seq to compare these models to their cells of mutation and to human datasets to elucidate the fundamental transcriptional programs and markers within and across these tumor types. This investigation will assess and credential these models against their clinical counterparts by scrutinizing the resulting datasets and validating their clinical relevance as pre-clinical models for therapeutic discovery and testing.