In support of the ERC open access strategy, it is critical for the ERC to identify published scientific articles that have been based on its funding schemes, in order to assess the impact of the research supported. Currently, the relationship between funding source and article is made only sporadically, years later in end-of-grant reports, or lost altogether, as there is no structured method for recording these associations. Ideally, the funding source should be identified as part of journal manuscript submission processes, however, this idea is currently in its very early stages and will not address the backlog of articles already published. Therefore, we propose to use text-mining methods to identify funding statements in full text articles available in Europe PubMed Central as an accurate and cost-effective solution to linking articles to funding sources. We will assess the outcomes through standard text-mining quality-assessment methods as well as through consultation with ERCEA staff. Anticipating useful outcomes, we will use the methods developed to identify ERC-funded articles in Europe PubMed Central on a routine basis, and make those results available in the search and article browse features on the Europe PubMed Central website. Finally, we will explore the feasibility of extending the approach to other Europe PubMed Central funders.

The quantitative understanding of the early development of mammalian embryos is essential to the progress of reproductive medicine. Yet, the physical and mechanical principles governing their morphogenesis remain largely unknown. Early mouse embryos self-organize by a succession of cell divisions, deformations and rearrangements, leading ultimately to the specification of two distinct cell lineages, segregated in inside and outside layers. Mechanical forces are therefore as important as biochemical activity in this process and precise 4-dimensional imaging of cells within the embryo reveals intense surface dynamics, regulated by contractile and adhesion proteins. However, our understanding of early embryos development still lacks a precise physical model integrating a dynamic description of the mechanical forces controlling cell shape and cell-cell adhesion. I will design a 4D physical model of the early mouse embryo providing accurate cell dynamics predictions. Cell shapes are primarily controlled by the actomyosin cortex and they will be described using recently developed cortical active shell theories. To represent accurately cell-cell adhesion dynamics, I will consider the crosstalk between cortical and adhesion proteins activities. Importantly, this model will be designed in close collaboration with an experimental group expert in the biophysical characterization of the mouse embryo, to incorporate measured mechanical parameters and molecular regulation mechanisms. Our model will be refined through cycles of theoretical predictions and experimental validations to uncover the principles of early mammalian embryos development and, more specifically, the mechanism of cell internalization at the 8 to 16 cells transition. This interdisciplinary project, at the interface between physical modeling and developmental biology will provide a unique and accurate biophysical framework for understanding the morphogenesis of early mammalian embryos.

Influenza is a major public health burden, with seasonal outbreaks contributing significantly to mortality worldwide, and the emergence of pandemic strains remaining an ever-present threat. Influenza drug and vaccine conception efforts are aided by a thorough understanding of its molecular biology. A key aspect of the influenza lifecycle is the production of capped and poly-adenylated messenger RNA by the heterotrimeric influenza polymerase (FluPol). Ground-breaking work performed by the Cusack lab, has described with residue-resolution detail, the FluPol structures that form during transcription of short, non-nucleoprotein (NP) bound viral RNAs (vRNAs). However, influenza transcription in vivo occurs within the ribonucleoprotein (RNP) particle and does not utilise naked genome segments. The viral RNP (vRNP) is a super-helical complex composed FluPol bound at the conserved 3′ and 5′ ends of a vRNA, which is coated with NP. The current low-resolution structures provide little information about the molecular details of vRNP function, particularly, how NPs interact with FluPol and the vRNA template. Via an inter-disciplinary approach, I will utilise cryo-electron microscopy methods, transcription assays and single-molecule fluorescence, to obtain the first high resolution structure of a dynamic influenza vRNP, with a particular focus on the spatial organisation of NPs relative to FluPol. In addition to this work facilitating future influenza drug research, it will provide a basis to investigate the vRNP during other lifecycle stages and act as proof-of-principle for study of other viral protein-RNA complexes, such as those from corona-, arena- and bunyaviruses. Work will be performed in the groups of Stephen Cusack and Olivier Duss based at EMBL Grenoble and Heidelberg, respectively. Here, I will have access to world-leading facilities and training opportunities, supporting my growth as an independent researcher and an expert in RNA virus structural biology.