We are in the midst of a revolution in our understanding of the internal organization of cells. In the 1950s we learned that lipid bilayer-based membranes serve as containers (organelles) within the cytoplasm. Now we are learning that liquid-like “membrane-less” organelles i.e. without any container, self-assemble based on “liquid-liquid” phase separations. We propose the seemingly radical idea that membrane-bounded organelles– like their membrane-less counterparts- are stabilized or even templated by analogous phase separations of their surface proteins into largely planar liquids akin to liquid crystals. Our unique Synergy team is organized specifically to test this “liquid crystal hypothesis” on the cell’s secretory compartments - ER exit sites (ERES) and the Golgi stack - by employing our complementary skills in physics, physical chemistry, biochemistry and cell biology. We hypothesize based on pilot experiments evidence that the ERES and Golgi self-organize as a multi-layered series of adherent liquid crystal-like phases of “golgin” and similar proteins which surround and enclose their membranes. Their differential adhesion and repulsion would specify the topology and dynamics of the membrane compartments. If this is true, it will literally rewrite the history of cell biology. We will test the ‘liquid crystal’ hypothesis directly, systematically, and quantitatively on an unprecedented scale to either modify/disprove it or place it on a firm rigorous footing. Experiments (Aim 1) with 13 pure golgins in cis and trans pairwise combinations will establish their foundational physical chemistry. Surgically engineered changes in golgins/ERES proteins will alter the rank order (hierarchy) of their affinities for each other and link phase separation physics to cell biology (Aim 2) and be used to establish the structural basis of phase separations and their specificity, and the potential for self-assembly of wholly synthetic biological organelles (Aim 3).

Artificial intelligence is rapidly permeating diverse domains. In the criminal procedural field, AI is enhancing the efficiency of law enforcement authorities and criminal justice systems in both preventing and prosecuting crimes, through the streamlining of tasks like data analysis, evidence assessment, and crime prediction. Despite this accelerating trend, a conspicuous gap persists due to the lack of a clear and comprehensive legal framework at both national and supranational levels. An additional factor hindering a responsible and beneficial implementation of AI in the criminal procedural sector is that some technical characteristics of the AI prototypes developed so far (e.g. opacity) interfere with some criminal procedural principles and standards. The lack of dialogue between legal scholarship and scientific community prevents the identification of suitable technical solutions. Given this scenario, managing the legal and ethical implications of AI in the criminal justice system proves to be a prohibitive challenge. The aim of FUTURE is twofold: to support national legislators across the EU in the regulation of artificial intelligence in criminal procedure, while fostering a harmonized approach to the subject; to bridge the gap between the technical and the legal community, ensuring the compliance of AI systems implemented in the criminal procedural field with its principles and safeguards. To achieve these objectives, FUTURE will firstly employ the legal comparison method in a groundbreaking manner: it will be applied on a comprehensive legal and empirical data set, covering all EU Member States’ jurisdictions. Secondly, it will bridge the expertise of legal and computer science professionals, giving them the first tangible chance to investigate the technical issues obstructing the use of AI in criminal proceedings and to assess the feasibility of technical solutions.

The Collaborative Economy (CE) is rapidly expanding through new forms of Internet labor and commerce, from Wikipedia to Kickstarter and Airbnb. However, it suffers from 3 main challenges: (1) Infrastructure: centralized surveillance that the central hubs of information exercise over their users, (2) Governance: disempowered communities which do not have any decision-making influence over the platform, and (3) Economy: concentration of profits in a few major players who do not proportionally redistribute them to the contributors. How can CE software platforms be implemented for solving these challenges? P2PMODELS explores a new way of building CE software platforms harnessing the blockchain, an emerging technology that enables autonomous agent-mediated organizations, in order to (1) provide a software framework to build decentralized infrastructure for Collaborative Economy organizations that do not depend on central authorities, (2) enable democratic-by-design models of governance for communities, by encoding rules directly into the software platform, and (3) enable fairer value distribution models, thus improving the economic sustainability of both CE contributors and organizations. Together, these 3 objectives will bootstrap the emergence of a new generation of self-governed and more economically sustainable peer-to-peer CE communities. The interdisciplinary nature of P2PMODELS will open a new research field around agent-mediated organizations for collaborative communities and their self-enforcing rules for automatic governance and economic rewarding. Bringing this proposal to life requires a funding scheme compatible with a high-risk/high-gain vision to finance a fully dedicated and highly motivated research team with multidisciplinary skills.

In nature there are many organisms able to control the ice nucleation rate of water. This ability allows such organisms to adapt to environmental changes, like large temperatures excursions, and to facilitate the search for food. Bio-molecules such as antifreeze proteins (AFPs) and ice nucleator proteins (INPs) are known to influence the ice nucleation rate, a feature that attracts great interest from a wide spectrum of scientific disciplines like biology and atmospheric science, and it offers several technological applications like cryo-preservation of tissues and increasing frozen food shelf life. I propose a new approach, based on a novel combination of water-protein coarse-grain model, able, for a wide range of temperatures and pressures, to deeply explore the configurational space of a water-protein solutions. The ProFrost project aims at defining a novel theoretical framework within which it will be possible to study how the folding properties of AFPs and INPs affect the thermodynamic state point of water. The main goal of the project is to understand the influence of protein interfaces on the dynamical and structural properties of water. The success of this project will pave the way for the computer based design of artificial functionalized protein sequences capable of influencing the phase of water. This project follows an innovative research line, with a multi-scale approach that combines multiple fields of research, such as physics, biology and chemistry. This research due to its interdisciplinary character and broad interest, has large impact and is often subject of publications on relevant international scientific journals, such as Nature, Science, Proceeding of National Academy of Science, Physical Review Letters. ProFrost will carried out under the supervision Prof Carlos Vega, a leading expert on water, and in collaboration with Prof. Dellago and Dr. Coluzza, expert in modelling and simulations of biophysical systems,