ERC FUNDED PROJECTS

Future data-driven modelling is increasingly challenging for many systems due to higher complexity levels, such as in energy systems, environmental and climate modelling, traffic and transport, industrial processes, health, safety, and others. This requires powerful concepts and frameworks that enable the design of high quality predictive models. In this proposal E-DUALITY we will explore and engineer the potential of duality principles for future data-driven modelling. An existing example illustrating the important role of duality in this context is support vector machines, which possess primal and dual model representations, in terms of feature maps and kernels, respectively. Within this project, besides using existing notions of duality that are relevant for data-driven modelling (e.g. Lagrange duality, Legendre-Fenchel duality, Monge-Kantorovich duality), we will also explore new ones. Duality principles will be employed for obtaining a generically applicable framework with unifying insights, handling different system complexity levels, optimal model representations and designing efficient algorithms. This will require taking an integrative approach across different research fields. The new framework should be able to include e.g. multi-view and multiple function learning, multiplex and multilayer networks, tensor models, multi-scale and deep architectures as particular instances and to combine several of such characteristics, in addition to simple basic schemes. It will include both parametric and kernel-based approaches for tasks as regression, classification, clustering, dimensionality reduction, outlier detection and dynamical systems modelling. Higher risk elements are the search for new standard forms in modelling systems with different complexity levels, matching models and representations to system characteristics, and developing algorithms for large scale applications within this powerful new framework.

2 492 500 €

Start date: 2018-10-01, End date: 2023-09-30

Our astonishing cognitive abilities are the consequence of complex connectivity within our neuronal networks and the large functional diversity of excitable nerve cells and their synapses. Investigations over the past half a century revealed dramatic diversity in shape, size and functional properties among synapses established by distinct cell types in different brain regions and demonstrated that the functional differences are partly due to different molecular mechanisms. However, synaptic diversity is also observed among synapses established by molecularly and morphologically uniform presynaptic cells on molecularly and morphologically uniform postsynaptic cells. Our hypothesis is that quantitative molecular differences underlie the functional diversity of such synapses. We will focus on hippocampal CA1 pyramidal cell (PC) to mGluR1α+ O-LM cell synapses, which show remarkable functional and molecular heterogeneity. In vitro multiple cell patch-clamp recordings followed by quantal analysis will be performed to quantify well-defined biophysical properties of these synapses. The molecular composition of the functionally characterized single synapses will be determined following the development of a novel postembedding immunolocalization method. Correlations between the molecular content and functional properties will be established and genetic up- and downregulation of individual synaptic proteins will be conducted to reveal causal relationships. Finally, correlations of the activity history and the functional properties of the synapses will be established by performing in vivo two-photon Ca2+ imaging in head-fixed behaving animals followed by in vitro functional characterization of their synapses. Our results will reveal quantitative molecular fingerprints of functional properties, allowing us to render dynamic behaviour to billions of synapses when the connectome of the hippocampal circuit is created using array tomography.

2 498 750 €

Start date: 2018-10-01, End date: 2023-09-30

We will reveal the neuronal mechanisms of fundamental hippocampal and axonal functions using direct patch clamp recordings from the small axon terminals of the major glutamatergic afferent and efferent pathways of the dentate gyrus region. Specifically, we will investigate the intrinsic axonal properties and unitary synaptic functions of the axons in the dentate gyrus that originate from the entorhinal cortex, the hilar mossy cells and the hypothalamic supramammillary nucleus. The fully controlled access to the activity of individual neuronal projections allows us to address the crucial questions how upstream regions of the dentate gyrus convey physiologically relevant spike activities and how these activities are translated to unitary synaptic responses in individual dentate gyrus neurons. The successful information transfers by these mechanisms ultimately generate specific dentate gyrus cell activity that contributes to hippocampal memory functions. Comprehensive mechanistic insights are essential to understand the impacts of the activity patterns associated with fundamental physiological functions and attainable with the necessary details only with direct recordings from individual axons. For example, these knowledge are necessary to understand how single cell activities in the entorhinal cortex (carrying primary spatial information) contribute to spatial representation in the dentate (i.e. place fields). Furthermore, because the size of these recorded axon terminals matches that of the majority of cortical synapses, our discoveries will demonstrate basic biophysical and neuronal principles of axonal signaling that are relevant for universal neuronal functions throughout the CNS. Thus, an exceptional repertoire of methods, including recording from anatomically identified individual small axon terminals, voltage- and calcium imaging and computational simulations, places us in an advantaged position for revealing unprecedented information about neuronal circuits.

1 994 025 €

Start date: 2018-04-01, End date: 2023-03-31

Evolution of resistance towards a single drug simultaneously increases (cross-resistance) or decreases (collateral sensitivity) fitness to multiple other antimicrobial agents. The molecular mechanisms driving cross-resistance are relatively well described, but it remains largely unclear how frequently does genetic adaptation to a single drug increase the sensitivity to others and what the underlying molecular mechanisms of collateral sensitivity are. This proposal focuses on studying the bacterial evolution of resistance and collateral sensitivity against antimicrobial peptides (AMPs). Beyond their modulatory roles in the immune system, these naturally occurring peptides provide protection against pathogenic microbes, and are considered as promising novel alternatives to traditional antibiotics. However, there are concerns that evolution against therapeutic AMPs can readily develop and as a by-product this might compromise natural immunity. Our knowledge of these issues is limited due to the shortage of systematic evolutionary studies. Therefore, the three central questions we address are: Do bacteria resistant to multiple antibiotics become hypersensitive to certain antimicrobial peptides? What are the evolutionary mechanisms leading to AMP resistance and to what extent does this process induce cross-resistance/collateral sensitivity against other drugs? Last, are these evolutionary trade-offs predictable based on chemical and functional peptide properties? To investigate these issues rigorously, we integrate tools of laboratory evolution, high-throughput phenotypic assays, functional genomics, and computational systems biology. Our project will provide an insight into the evolutionary mechanisms that drive cross-resistance and collateral sensitivities with the aim to explore the vulnerable points of resistant bacteria. Another goal is to provide guidelines for the future design of antimicrobial peptides with desirable properties against bacterial pathogens.

1 846 250 €

Start date: 2015-10-01, End date: 2021-09-30