Antibiotic resistance (AMR) is a major public health issue, with 5 million deaths in 2019 linked to AMR worldwide. These numbers are comparable to the toll of the SARS-CoV-2 pandemic. Without new solutions, AMR is projected to soon become one of the leading causes of death in the EU. Addressing this challenge requires the development of new, effective antibiotics, but this alone is not sufficient due to the rapid evolution of bacteria. Understanding the drivers and mechanisms of AMR is vital to delay or reverse resistance in both existing and new antibiotics, especially since no new broad-spectrum antibiotics have been developed since the 1990s and their development is a lengthy process with high attrition rates. The ENDAMR doctoral network aims to better equip researchers in Europe to understand and develop new strategies to tackle AMR. WP1 focuses on how AMR affects the fitness of pathogenic bacteria in the gut microbiome, aiming to identify microbiome characteristics that predispose to AMR infections and to explore microbiome-based interventions. WP2 examines AMR acquisition via horizontal gene transfer, investigating evolutionary pathways, host genetics, environmental factors, dissemination, and AMR reservoirs. WP3 is dedicated to understanding the frequency, mechanisms, and clinical implications of antibiotic resistance, with a particular focus on the less studied aspects of heteroresistance and tolerance, and to developing diagnostic tools and predictive models. WP4 explores the combination of antibiotics to enhance treatment outcomes and potentially prevent or reverse AMR, based on understanding the interplay of resistance mechanisms. ENDAMR will also prepare doctoral candidates for various career paths beyond academia, including teaching, science communication, and entrepreneurship. Candidates will gain transferable skills and learn from industry role models, equipping them to make significant contributions to solving the AMR crisis.

Here I propose an ambitious project designed to explore an exciting new hypothesis: TFs are non-randomly spatially organised within the cell nucleus, forming radial gradients that likely play an important role in regulating gene expression. This idea was born from one of the intriguing discoveries revealed by Genomic loci Positioning by Sequencing (GPSeq), a technique developed in the host lab, that the binding motifs of the majority of human transcription factors (TFs) seem to form radial density gradients, with some TF binding motifs being more abundant towards the nuclear periphery, whilst others more abundant in the centre. Furthermore, preliminary analysis showed that the predicted binding motifs of master regulator TFs specifying lymphoid vs myeloid lineages form opposing radial gradients, suggesting that this mode of transcriptional regulation is critical for cell fate specification. In this project, I will differentiate hematopoietic stem cell (HSC) into opposing lineages (lymphoid/myeloid) and perform TF CUT&RUN to assay TF occupancy and chromatin landscape, coupled with GPSeq to assay the radial arrangement of chromatin in these lineages. The intersection of these two genome-wide datasets will provide the first ever map of the occupancy of lineage specific transcription factors through differentiation time and nuclear space with respect to radiality. In parallel I will generate additional omics datasets in these cell populations to assay chromatin accessibility, transcription and chromatin interactions to build a comprehensive, multi-dimensional genomic atlas in which the spatial arrangement of TFs can be linked with gene expression dynamics. Crucially, no other study has thus far addressed this novel hypothesis; my work will provide significant advancements in understanding the interplay of genome architecture with transcriptional regulation and cell-fate specification.

Machine learning (ML) has enabled and accelerated frontier research in the life sciences, but democratised access to such methods is, unfortunately, not a given. Access to necessary hardware and software, knowledge and training, is limited, while methods are typically insufficiently documented and hard to find. Furthermore, even though modern AI-based methods typically generalize well to unseen data, no standard exists to enable sharing and fine-tuning of pretrained models between different analysis tools. Existing user-facing platforms operate entirely independently from each other, often failing to comply with FAIR data and Open Science standards. The field of AI and ML is developing at a staggering pace, making it impossible for the non-specialist to stay up to date. To enable the life science communities to benefit from AI/ML-powered image analysis methods, AI4LIFE will build bridges, providing urgently needed services on the common European research infrastructures. We will build an open, accessible, community-driven repository of FAIR pre-trained AI models and develop services to deliver these models to life scientists, including those without substantial computational expertise. Our direct support and ample training activities will prepare life scientists for responsible use of AI methods, while contributor services and open standards will drive community contributions of new models and interoperability between analysis tools. Open calls and public challenges will provide state-of-the-art solutions to yet unsolved image analysis problems in the life sciences. Our consortium brings together AI/ML researchers, developers of popular open source image analysis tools, providers of European-scale storage and compute services and European life sciences Research Infrastructures -- all united behind the common goal to enable life scientists to fully benefit from the untapped but potentially tremendous power of AI-based analysis methods.

Cilia-AI will train a new generation of multidisciplinary biomedical researchers and entrepreneurs, and those specializing in emerging machine learning technologies, a subset of AI. The focus is the study of primary cilia, microtubule-based projections on cell surfaces that play a pivotal role in coordinating cellular signalling pathways during development and homeostasis of cells, tissues and organs. These tiny structures are essential for various physiological functions such as hearing, smell, respiration, excretion and reproduction. Dysfunctional cilia can lead to >35 severe human diseases known as ciliopathies, exhibiting diverse and overlapping phenotypes, affecting up to 1 in 400 people. To unravel the multi-level organisation and regulation of cilia in health and disease, Cilia-AI employs a multidisciplinary approach, integrating cutting edge technologies like structural biology, omics- and organoid technologies. Advanced imaging techniques, including super-resolution microscopy, cryo-electron tomography and expansion microscopy, will be used to generate high-resolution and versatile datasets. Processing such data requires sophisticated computational methods. Cilia-AI is at the forefront of implementing and developing machine learning approaches to decipher these high-content datasets and integrate diverse multidisciplinary data. Cilia-AI offers unparalleled training opportunities for 14 Doctoral Candidates (DCs) in both academic and industrial settings. The training involves individual research projects, secondments, and network-wide sessions. This training equips DCs with skills attractive to both industrial and academic sectors, enhancing their career prospects in these domains. Overall, Cilia-AI’s research and training activities contribute to advancing the understanding of cilia in health and disease while fostering a new generation of skilled professionals with broad competences.

Metabolic functions in vertebrates are ruled by thyroid hormones, produced by the thyroid gland and secreted into the bloodstream. Thyroid hormone levels modulate growth, development and many other key processes across the lifespan, and their impairment severely affects human health. Despite extensive studies, controlling hormone levels in vivo remains a major challenge. Synthesis of thyroid hormones occurs extracellularly, via iodination of thyroglobulin. They must then undergo endocytosis before they can be released into the bloodstream. Yet, most thyroglobulin forms a colloid matrix within the gland, enabling iodine and hormone storage for the body. How these three mechanisms are balanced to provide physiological levels of hormones to the organism remains an outstanding question. Using an integrative structural biology approach, the proposed research will focus on unravelling the molecular mechanisms underlying the major checkpoints of thyroid hormone synthesis. My research group will combine electron cryo-microscopy with biochemical and light microscopy tools to study thyroid hormone regulation at different scales: from in vitro reconstituted systems to native thyroid organoids. First, we will elucidate how thyroglobulin iodination modulates hormone yields and modifies thyroglobulin structure. Second, we will determine the mechanism of thyroglobulin endocytosis regulated by two main receptors. Third, we will decipher the mechanism of storage and depletion of the thyroglobulin storage matrix. These approaches will allow us to reproduce the complexity of thyroid hormone synthesis at high resolution and in a native environment. My expertise in structural biology and my pioneering work on thyroglobulin puts me in an ideal position to carry out this work. The expected outcome and method development of this research will open a new front in thyroid structural biology and inspire alternative strategies for the control of thyroid hormone synthesis in vivo.