ABSTRACT Soft-tissue sarcomas are heterogeneous tumors that originate from cells belonging to the mesenchymal lineages, and that affect almost 200,000 individuals worldwide each year. The most aggressive and metastatic sub-types in adults are those with complex karyotypes and multiple genetic aberrations. The overall survival of these sarcoma patients has not greatly improved in recent years, and alternative approaches to chemotherapy and radiotherapy such as immunotherapy have so far provided only marginal benefits. Novel therapeutic advances for UPS are hindered by the lack of knowledge about the functional consequences of the complex genomic alterations found in patients and limited characterizations of the tumor microenvironment (TME), which would reveal non-cell-autonomous mechanisms critical to sarcoma progression. Experimental tools and available animal models currently do not address these limitations. However, appropriate models could facilitate the efficient discovery of new targets and immune- based therapies for these tumors, which have relatively low incidence and for which the development of clinical trials is often challenging. Accordingly, we propose to generate sarcoma mouse models that encompass the actual somatic aberrations observed in patients. Moreover, we will use these models to facilitate studies of treatment response and TME composition. The employment of these new models, together with newer technologies such as single-cell RNA- sequencing (scRNA-seq), CyTOF and Imaging Mass Cytometry (IMC) will ultimately illuminate the key expression profile of the single tumor cells, mechanisms of metastasis, resistance to conventional treatments and TME components that may influence such mechanisms. Ultimately, these models will translate to the clinic more effective therapeutic combinations and regimens. Successful completion of this project will i) generate new mouse models of complex sarcoma that recapitulate the genetic defects found in human sarcoma and provide a comprehensive functional characterization of these models (Aim 1), ii) illuminate the expression profile of the tumor cells and discrete sub-groups of them, to understand how these profiles influence the TME composition (Aim 2), iii) test how these models respond to different radiotherapy administration schedules in the heterogeneous settings of distinct tumor genetics, expression profiles and environmental elements (Aim 3).