Predicting clinical response to novel and existing anticancer drugs remains a major hurdle for successful cancer treatment. Studies indicate that the tumor ecosystem, resembling an organ-like structure, can limit the predictive power of current therapies that were evaluated solely on tumor cells. The interactions of tumor cells with their adjacent microenvironment are required to promote tumor progression and metastasis, determining drug responsiveness. Such interactions do not form in standard research techniques, where cancer cells grow on 2D plastic dishes. Hence, there is a need to develop new cancer models that better mimic the physio-pathological conditions of tumors. Here, we create 3D-bioprinted tumor models based on a library of hydrogels we developed as scaffold for different tumor types, designed according to the mechanical properties of the tissue of origin. As PoC, we bioprinted a vascularized 3D brain tumor model from brain tumor cells co-cultured with stromal cells and mixed with our hydrogels, that resemble the biophysics of the tumor and its microenvironment. Our patient-derived models consist of cells from a biopsy, constructed according to CT/MRI scans, and include functional vessels allowing for patients' serum to flow when connected to a pump. These models will facilitate reproducible, reliable and rapid results, determining which treatment suits best the specific patient's tumor. Taken together, this 3D-printed model could be the basis for potentially replacing cell and animal models. We predict that this powerful platform will be used in translational research for preclinical evaluation of new therapies and for clinical drug screening, which will save critical time, reduce toxicity and significantly decrease costs generating a major societal benefit. Our platform offers a highly attractive business case, as pharmaceutical and biotech companies heavily invest in preclinical predictive tools for novel personalized drug screening strategies.

When you look at a given object - are you calculating its value even if not prompted to doWhen you look at a given object - are you calculating its value even if not prompted to do so? Most theories of valuation claim that choice, prompt or in general elicitation are needed to induce valuation. However, the well-known mere exposure effect dating back to 1968, suggests that merely viewing an item can induce valuation and even enhance value. In this proposal I aim to demonstrate that valuation is an early and automatic process relying on visual, attentional and motor systems. PassiveValueMarkers offers a novel framework to identify biomarkers of value of individual items passively, without elicitation. The aims of PassiveValueMarkers are: 1) Identifying behavioural and neural passive markers for individual items using computational modelling in and influencing these markers ; 2) Detecting passive markers of value in gaze pattern analysis behaviourally and in the brain using fMRI; 3) Considering the gap that exists between laboratory studies and the real world, I will identify passive value markers in naturalistic virtual reality environments. The overreaching aim of this proposal is to develop a new theoretical framework on individual passive value construction and change. To do so, I will use a unique combination of neuroimaging, computational modelling, gaze-tracking analysis and virtual reality. This research will directly address an understudied area of how value for individual items is formed by the brain without active prompt or elicitation. Uncovering the mechanisms of passive value representation at the single item level and individualized per participant, will allow to design closed loop manipulations at the item level. This approach will serve as the basis for developing novel evidence-based methods for enhanced preference modification in healthy participants and in disorders with abnormal valuation such as addictions, mental illness, and eating disorders.

What is the origin of the electromagnetic (EM) counterparts of gravitational waves observed from compact binary mergers? What makes short gamma ray bursts (GRBs)? What are the sources of IceCube’s high-energy neutrinos? Are all core-collapse supernovae exploding via the same mechanism? These are some of the puzzles that have emerged with the rapid progress of time domain astronomy. Relativistic jets in compact binary mergers and GRBs, and their interaction with the surrounding media hold the key to these, and other, seemingly unrelated broad-impact questions. Here I propose a new forefront study of how relativistic jets interact with their surrounding media and of its numerous implications, focusing on compact binary mergers and GRBs. The goal of this project is to study, first, the jet-media interaction, and the microphysics of the radiation-mediated shocks that it drives. I will then use the results, together with available observations, to learn about compact binary mergers, GRBs and SNe, sheding light on the questions listed above, and probing the nature of relativistic jets in general. Important goals will include: (i) General models for the propagation of relativistic jets in various media types. (ii) Modeling of the EM signal generated by jet-media interaction following compact binary mergers. (iii) Estimates of the neutrino signal from jet-media interaction in GRBs and SNe. (iv) Constraint the role of jets in SN explosions. This project is timey as it comes at the beginning of a new multi-messenger era where the EM counterparts of GW sources are going to be detected on a regular basis and where the face of transient astrophysics is going to be changed by a range of large scale surveys such as LSST, the SKA, and more. This project will set the theoretical base for understanding numerous known and yet-to be discovered transients that will be detected in the next decade.

GrowBot proposes a disruptively new paradigm of movement in robotics inspired by the moving-by-growing abilities of climbing plants. Plants are still a quite unexplored model in robotics and ICT technologies, as their sessile nature leads to think that they do not move. Instead, they move greatly, on a different time scale, purposively, effectively and efficiently. To move from one point to another, plants must grow and continuously adapt their body to the external environmental conditions. This continuous growth is particularly evident in climbing plants. By imitating them, the GrowBot objective is to develop low-mass and low-volume robots capable of anchoring themselves, negotiating voids, and more generally climbing, where current climbing robots based on wheels, legs, or rails would get stuck or fall. Specifically, the ability to grow will be translated by additive manufacturing processes inside the robot, which creates its body by depositing new materials with multi-functional properties, on the basis of the perceived external stimuli (without a pre-defined design). Energy efficiency will be intrinsic to such approach, but novel bio-hybrid energy harvesting solutions will be also implemented to generate energy by interfacing soft technologies with real plants. Perception and behavior will be based on the adaptive strategies that allow climbing plants to explore the environment, described mathematically after experimental observations. GrowBot would contribute to consolidate this ground-breaking and pioneering research area on plant-inspired robotics that, although still in its infancy, can represent a revolutionary approach in robotics, as it has already happened with plant-inspired solutions in material science. GrowBot is based on a strongly interdisciplinary character and can open the way for a new technological paradigm around the concept of growing robots, fostering a European innovation eco-system for several high-tech sectors.