ARISE2

The sequence families team develops the InterPro, Pfam, Rfam and RNAcentral databases. This provides and excellent environment to learn and help develop cutting edge information resources. At present we are rapidly adopting AI technologies to assist in coding and curation. There are many opportunities and we are increasing our ambitions to provide detailed and accurate biological information more rapidly to the community. Projects could focus on using AI to improve data, or the software infrastructure for any of the data resources we provide.

Our team develops and applies Biological Small-Angle X-ray Scattering (BioSAXS) methods to support structural biology research, combining experimental development, beamline operation, and user service provision. We advance SAXS methodologies, including high-throughput and complex sample-environment approaches, while continuously improving beamline infrastructure, data workflows, and user-facing services. Current and future developments may involve both computational and experimental directions, such as data-processing pipelines, databases, automation strategies, or innovative sample environments. Fellows will have the opportunity to contribute according to their expertise, whether in computational method development, experimental instrumentation, sample-environment innovation, or integrated service provision within a collaborative and multidisciplinary beamline environment.

The EMBL Rome Light Imaging Facility is seeking ARISE2 postdoctoral fellows to develop automated imaging technology and image analysis for large-scale biological research. We develop technology that transforms service provision in imaging-based spatial omics as well as general areas of fluorescence microscopy and image analysis. Possible projects include “lab-in-the-loop” approaches for smart microscopy, robotics, and automation, as well as software development pipelines for image analysis.

Our current work in X-ray imaging focuses on the development of X-ray imaging as part of a multimodal and multi-lengthscale workflow. Amongst other things, we wish to link together lab-based X-ray Imaging with our synchrotron beamline-based setup. We work closely with users of other biological imaging modalities to link together, for example, X-ray imaging with volume electron microscopy. Typically, our samples range in size from 0.5mm – 4mm and come in multiple forms – fully hydrated, mounted in liquid, formalin fixed in paraffin, or heavy metal stained and resin embedded, and the data are collected at room temperature. Of interest is extending the techniques we offer to include data collection on fresh frozen tissue samples.

Our research develops and optimises epigenome editing technologies as a scientific infrastructure for discovery and translation. Epigenome editing enables precise, reversible modulation of gene expression without permanent DNA sequence alterations, providing powerful research tools for dissecting genome regulation and disease.

EMBL – EBI’s imaging databases, the BioImage Archive and EMPIAR (Electron Microscopy Public Image Archive) are used by tens of thousands of scientists, who provide and access life sciences images. They have strong interactions with both EMBL’s world-leading imaging facilities, as well as the worldwide BioImaging community. This gives us the ideal position to develop image data conversion, storage, compression, visualisation, and analysis technologies. All our technology development is primarily applied to improve our service offering to imaging scientists, image analysts, and researchers in the biological sciences. Automation of FAIR data management, particularly through AI processes (e.g., LLM agents, MCP servers ), and application to multimodal data (e.g., spatial transcriptomic) is a key strategic goal for our service development.

The EMBL Rome Cornelius Gross Group is seeking ambitious ARISE2 postdoctoral fellows to join a technology-focused research program aimed at transforming the large-scale behavioral monitoring of laboratory animals. The project will develop integrated next-generation platforms combining continuous live video monitoring, RFID-based tracking, custom electronics, scalable data storage architectures, and AI-driven behavioral analysis to enable high-throughput, longitudinal phenotyping in socially housed animals. A major focus will be the development of “lab-in-the-loop” approaches in which machine learning models dynamically guide iterative phenotypic assessment and intervention, creating adaptive experimental systems for studying brain function, behavior, and disease in ethologically relevant settings.

Our team develops technologies that help researchers discover, access, and use biological data and scientific knowledge. We maintain EBI Search, EMBL-EBI’s central discovery platform, and develop emerging AI-driven services such as DocBot. Current work focuses on natural language search, AI agents, large language models, and semantic discovery technologies that make complex scientific resources easier to find, understand, and use.

My group develops machine learning-based image analysis methods and tools. Our tools aim to democratize access to cutting-edge AI technology for biologists without a professional computational background. We collaborate extensively with other developers of user-facing tools in projects such as AI4Life, building the future of FAIR AI model exploration and sharing. On the method development side, we are interested in all aspects of user-friendly deep learning, from interactive tools to reusable foundational models. On the infrastructure side, we collaborate with Wei Ouyang (SciLifeLab) and Matthew Hartley (EMBL-EBI) to deliver the BioImage Model Zoo and associated services for pre-trained model and training dataset selection.

Projects could include method development and software engineering to scale up genomic/protein analyses to full databases, work to create smaller versions of models for deployment in low-resource settings, or expansion of methods from single species to pan-pathogen applications. Example topics are pangenomes, serotyping, anti-microbial resistance, and genomic surveillance. Our work is undertaken in collaboration with computational service teams at EMBL-EBI, such as MGnify, Ensembl, and UniProt. 

Our group is developing scalable technologies to modernise expert biocuration, moving from predominantly manual workflows towards AI-assisted curation while maintaining UniProt’s high standards of quality, reliability, and scientific accuracy. Current and future work focuses on developing human-in-the-loop AI models and curation tools that can extract, structure, and validate biological knowledge from the literature and other data sources at scale. An ARISE2 fellow would contribute to the development of specialised AI models for vaccine capsular antigen curation, enabling relevant antigen information to be captured, standardised, and delivered through UniProt for use by the wider research community. This technology will support FAIR data management by converting unstructured biological knowledge into structured, traceable, interoperable, and reusable annotations, with community feedback used to refine the models and improve service provision for external researchers.

The Marquez Team has pioneered the development of Online Crystallography; fully automated protein-to-structure pipelines integrating crystallization, synchrotron data collection, and crystallographic data analysis into continuous workflows operated via the web. These pipelines rely on the CrystalDirect technology and the Crystallography Data Management system and have been fully integrated with automated data collection at the ESRF synchrotron, enabling rapid structure determination and large-scale X-ray-based fragment screening. Our facility generates large amounts of AI-ready data, and we are now extending our work to introduce Machine Learning (ML) approaches for AI-driven automation and for hit-to-lead optimisation and structure-based drug design. Our interdisciplinary team offers opportunities for scientists, engineers, and software developers to work in one of the leading infrastructures for structural biology within the areas of protein crystallography, fragment screening, drug design, automation, and large-scale scientific data management and Machine Learning. Currently, we are particularly interested in profiles in structural biology or computer science oriented towards one or several of the following areas: fragment screening, structure-guided drug design, machine learning, and artificial intelligence.

The EMBL Grenoble and ESRF are keen to leverage the unique X-ray characteristics of the ESRF-EBS synchrotron source for next-generation biological X-ray imaging applications. These span from whole organ imaging on BM19, room temperature cellular and sub-micron resolution capabilities on a new experimental setup to be established on ID30B, and all the way to nano resolution X-ray imaging on ID16A and a new dedicated nano-bioimaging beamline on ID18 with both room temperature and cryogenically preserved biological samples. This ARISE project centers on developing and deploying a new sub-micron imaging service on the ESRF-EMBL jointly operated beamline ID30B at the ESRF. We also propose to develop a new generation HiTT instrument that is capable of acquiring sub-micron X-ray tomograms from large numbers of cryogenically preserved samples. With many future synchrotron upgrades to 4th generation source in various stages of planning and implementation, we believe f such new scientific capabilities would be timely for the growing biological X-ray imaging and CryoET community.

The Electron Microscopy Data Bank manages the public repository for electron microscopy maps and representative tomograms of macromolecular complexes and subcellular structures. We develop technologies and services that facilitate open access to published cryoEM data, add value by revealing facets of the data we hold, connections to complementary databases (including but not limited to EMPIAR and PDB), and we use our work to prepare the archive for future experimental technologies.

Our team develops tools and resources that help record, analyse, and interpret the biological effects of compounds, typically small molecules. The ChEMBL database of bioactivities extracted from the literature is widely used to support drug discovery in academia as well as throughout the pharmaceutical industry; ChEBI is an ontology and database of chemicals of biological interest whose identifiers are widely used by other biological databases; SureChEMBL is a database of molecules and biological targets extracted from the patent literature via a fully automated process. Our team is a mixture of researchers, data scientists, software engineers, and data curators. The work within the group ranges from curation of biological assay descriptions and chemical structures, applying NLP and AI approaches to extract relevant information from unstructured text, through to web development with modern technologies. Team members benefit from working in an open and collaborative environment, which, in particular, for an ARISE fellow, provides the opportunity to grow both in knowledge of service provision and technology development. The close alignment of our services with the pharmaceutical industry will also be of benefit.

The Papp Team at EMBL Grenoble develops scientific instrumentation for macromolecular X-ray Crystallography and Cryo-Electron Microscopy applications. One of our most recent and significant development projects is the EasyGrid technology, a system designed to fully automate sample preparation and quality control for cryo-imaging experiments. Pilot measurements demonstrated that the EasyGrid system is suitable for preparing samples for Single Particle Analysis experiments and showed the superiority of EasyGrid-prepared samples in terms of ice quality for Cryo-Electron Tomography and X-ray Nano Imaging techniques, compared to traditional plunge freezing equipment. Based on these solid technical developments, the team is now working on creating software tools to fully automate the Cryo-EM/ET sample preparation workflow, including an AI-based system for preparation parameter optimization using an on-the-fly sample quality control method.

We have developed a framework to allow genome-wide association testing for CNVs to be performed at scale in large human cohorts and have developed a set of guidelines for standardisation of CNV GWAS models. We will work closely with the GWAS catalog to implement infrastructure that is acutely needed to allow the long-term capture of CNV GWAS results under FAIR guidelines and the linkage of information across variant types (i.e., SNPs and CNVS). We will continue development of CNV GWAS methods and also apply the existing standard CNV GWAS models to large human cohort data. This will generate larger volumes of CNV GWAS results that can be directly deposited into the GWAS catalog and will drive the development of infrastructure for data linkage and additional CNV-specific tooling and functionality within the GWAS catalog

Our group develops retrieval-augmented and agentic AI systems for scientific discovery, with a focus on how AI can retrieve, organize, reason over, and interact with large collections of scientific knowledge and biological data. We build scalable and trustworthy AI technologies that support literature-grounded reasoning, hypothesis generation, experimental planning, and interaction with scientific tools and infrastructures. A central goal is to develop open and reusable AI services for the broader life-science community.

The EMBL Rome Flow Cytometry Facility offers high-end flow cytometry and cell sorting solutions. These tools are crucial for accurate single-cell data acquisition and for purifying cells for diverse downstream applications. We are seeking ambitious ARISE2 postdoctoral fellows to join a technology-focused and highly collaborative research program aimed at the development of scalable cytometry-based phenotyping and cell sorting workflows for Lifecourse, including conventional and imaging flow cytometry technologies and protocols, longitudinal immune profiling strategies, optimization of multiplex cytokine and flow cytometry panels, bulk- and single-cell sorting protocols, standardized sample processing pipelines, and integration of immune datasets with broader multi-omics and physiological readouts.

We are developing advanced optical imaging methods that are based on multi-photon microscopy, active wave-front shaping, photo-acoustics, and high-resolution spectroscopy. Our aim is to establish our new approaches as disruptive technologies in the life sciences and to further engineer and automate our prototypes for routine service provision.

The Saez-Rodriguez group at EMBL-EBI specializes in developing computational methods to understand the deregulation of cellular networks in diseases such as cancer, autoimmune disorders, and fibrosis. By integrating multi-omics data—including single-cell and spatial omics—with molecular interactions knowledge, the group creates mechanistic, context-specific models. These models aid in deciphering complex biological processes and are shared as open-source tools, facilitating their application by external researchers.

Our group is developing enabling technologies that push the technical limits in spatial biology research and applying them for understanding how molecular and spatial composition of cells relates to their phenotype. We have set up cutting-edge methods that integrate advanced imaging and omics (SABER, Light-Seq) down to subcellular spatial omics, and implement automated workflows for hardware-software communication for adaptive feedback microscopy. We are particularly interested in spatial organization of subcellular compartments and their transcriptomic and proteomic composition. We have recently developed a new in situ proximity detection approach (ProPER) and would like to develop new applications for high-throughput in situ annotation of molecular states of RNA and proteins, high-fidelity detection of small and/or surface RNAs, as well as cell-cell interactions. We would also incorporate this approach together with other commercial and custom spatial omics methods.

Our team builds and maintains the BioStudies database – a resource that facilitates transparent, reproducible science by aggregating and publishing all outputs of a scientific study. BioStudies acquires data via a variety of routes, both pre- and post-publication. We are looking to extend our infrastructure and enable application of data harmonisation methods and advances in AI to support new, emerging fields, while maintaining the generic nature of our services.

Volume electron microscopy (vEM) is a fast-evolving imaging field that aims at acquiring ultrastructural details of cells, small model organisms, and tissues in three dimensions. vEM generally produces large datasets that can reach multiple TB of data, for which extracting valuable morphometric information by image segmentation is still a challenge. For targeted studies, which restrict measurement to selected regions of interest, other imaging modalities (light microscopy or X-ray microscopy) are used to build volumetric maps of the samples in which sub-regions of interest are defined. Due to the size and complexity of such multi-modal image datasets, it is challenging to efficiently (ideally automatically) extract biophysical information from these information-rich datasets. To tackle this challenge, the Bioimage Analysis Support Team and Electron Microscopy Core Facility from EMBL Heidelberg, as well as the VIB Bio Imaging Core from KU Leuven, are teaming up to develop and deploy reusable computational workflows for the automated management and AI-based analysis of such correlative volume EM datasets.

We develop computational technologies to understand the physical principles governing living systems, combining mechanistic modelling, simulation, and data analysis. Our work includes software platforms for tissue and multicellular simulations as well as quantitative methods to extract biological insight from complex datasets. We aim to make these technologies accessible, reproducible, and FAIR, enabling researchers to efficiently analyse data, build predictive models, and integrate computational approaches into their research workflows.

The Protein Data Bank in Europe (PDBe) team develops essential macromolecular structure resources and tools for biologists and other life scientists. As a founding partner of the Worldwide Protein Data Bank organisation, we work on the Protein Data Bank (PDB). We also manage the community-led PDBe – Knowledge Base resource and the AlphaFold Protein Structure Database, a collaboration with DeepMind. Across these resources, PDBe’s work focuses on making macromolecular structure data accessible, integrated, and interpretable through advanced tools for data analysis, querying, and visualisation, supporting fundamental and translational research, education, and training.

Improving the functionality of the EMBL-EBI resources PRIDE and MetaboLights, and the integration of proteomics/metabolomics data with other omics data types (implementing the FAIR data principles) are two key aspects for the team in the near future. This offers the possibility for the fellow to work in different topics depending on their background (e.g., data management, data analysis, data visualisation, software development and infrastructure, Artificial Intelligence, etc). In the context of data integration for proteomics data, as mentioned above, this involves different data types such as gene and protein expression information (together with Expression Atlas and Open Targets), post-translational modifications (UniProt), (meta)proteomics data, and (meta)genomics sequences (Ensembl, MGnify), and also structural proteomics data such as crosslinking (PDBe). In the context of metabolomics, integration of metabolomics with other data types (as mentioned above) and providing improved support for exposomics data, the generation of public spectral libraries and interaction with ChEBI and other relevant EMBL-EBI resources are topics of high interest.

The instrumentation development team in the Scientific Instrumentation Workshops develops custom hardware platforms that enable new experimental capabilities for researchers at EMBL. The team is tightly integrated with our mechanical and electronics workshops, allowing quick prototyping and development cycles for even complex instrumentation projects. We have strong expertise in integrated electro-mechanical technologies that enhance sample handling and measurement workflows, and our current focus areas include: high-throughput (micro)fluidic handling; integration of fluid measurement cells into light and x-ray microscopy; and sample handling for cryo-EM. We are available as a collaboration partner across all facets of life science research at EMBL.

The EMBL Rome Light Imaging Facility is seeking ARISE2 postdoctoral fellows to develop automated imaging technology and image analysis for large-scale biological research. We develop technology that transforms service provision in imaging-based spatial omics as well as general areas of fluorescence microscopy and image analysis. Possible projects include “lab-in-the-loop” approaches for smart microscopy, robotics, and automation, as well as software development pipelines for image analysis.

Our current work in X-ray imaging focuses on the development of X-ray imaging as part of a multimodal and multi-lengthscale workflow. Amongst other things, we wish to link together lab-based X-ray Imaging with our synchrotron beamline-based setup. We work closely with users of other biological imaging modalities to link together, for example, X-ray imaging with volume electron microscopy. Typically, our samples range in size from 0.5mm – 4mm and come in multiple forms – fully hydrated, mounted in liquid, formalin fixed in paraffin, or heavy metal stained and resin embedded, and the data are collected at room temperature. Of interest is extending the techniques we offer to include data collection on fresh frozen tissue samples.

EMBL – EBI’s imaging databases, the BioImage Archive and EMPIAR (Electron Microscopy Public Image Archive) are used by tens of thousands of scientists, who provide and access life sciences images. They have strong interactions with both EMBL’s world-leading imaging facilities, as well as the worldwide BioImaging community. This gives us the ideal position to develop image data conversion, storage, compression, visualisation, and analysis technologies. All our technology development is primarily applied to improve our service offering to imaging scientists, image analysts, and researchers in the biological sciences. Automation of FAIR data management, particularly through AI processes (e.g., LLM agents, MCP servers ), and application to multimodal data (e.g., spatial transcriptomic) is a key strategic goal for our service development.

My group develops machine learning-based image analysis methods and tools. Our tools aim to democratize access to cutting-edge AI technology for biologists without a professional computational background. We collaborate extensively with other developers of user-facing tools in projects such as AI4Life, building the future of FAIR AI model exploration and sharing. On the method development side, we are interested in all aspects of user-friendly deep learning, from interactive tools to reusable foundational models. On the infrastructure side, we collaborate with Wei Ouyang (SciLifeLab) and Matthew Hartley (EMBL-EBI) to deliver the BioImage Model Zoo and associated services for pre-trained model and training dataset selection.  

The EMBL Grenoble and ESRF are keen to leverage the unique X-ray characteristics of the ESRF-EBS synchrotron source for next-generation biological X-ray imaging applications. These span from whole organ imaging on BM19, room temperature cellular and sub-micron resolution capabilities on a new experimental setup to be established on ID30B, and all the way to nano resolution X-ray imaging on ID16A and a new dedicated nano-bioimaging beamline on ID18 with both room temperature and cryogenically preserved biological samples. This ARISE project centers on developing and deploying a new sub-micron imaging service on the ESRF-EMBL jointly operated beamline ID30B at the ESRF. We also propose to develop a new generation HiTT instrument that is capable of acquiring sub-micron X-ray tomograms from large numbers of cryogenically preserved samples. With many future synchrotron upgrades to 4th generation source in various stages of planning and implementation, we believe f such new scientific capabilities would be timely for the growing biological X-ray imaging and CryoET community.

The Papp Team at EMBL Grenoble develops scientific instrumentation for macromolecular X-ray Crystallography and Cryo-Electron Microscopy applications. One of our most recent and significant development projects is the EasyGrid technology, a system designed to fully automate sample preparation and quality control for cryo-imaging experiments. Pilot measurements demonstrated that the EasyGrid system is suitable for preparing samples for Single Particle Analysis experiments and showed the superiority of EasyGrid-prepared samples in terms of ice quality for Cryo-Electron Tomography and X-ray Nano Imaging techniques, compared to traditional plunge freezing equipment. Based on these solid technical developments, the team is now working on creating software tools to fully automate the Cryo-EM/ET sample preparation workflow, including an AI-based system for preparation parameter optimization using an on-the-fly sample quality control method.

The EMBL Rome Flow Cytometry Facility offers high-end flow cytometry and cell sorting solutions. These tools are crucial for accurate single-cell data acquisition and for purifying cells for diverse downstream applications. We are seeking ambitious ARISE2 postdoctoral fellows to join a technology-focused and highly collaborative research program aimed at the development of scalable cytometry-based phenotyping and cell sorting workflows for Lifecourse, including conventional and imaging flow cytometry technologies and protocols, longitudinal immune profiling strategies, optimization of multiplex cytokine and flow cytometry panels, bulk- and single-cell sorting protocols, standardized sample processing pipelines, and integration of immune datasets with broader multi-omics and physiological readouts.

We are developing advanced optical imaging methods that are based on multi-photon microscopy, active wave-front shaping, photo-acoustics, and high-resolution spectroscopy. Our aim is to establish our new approaches as disruptive technologies in the life sciences and to further engineer and automate our prototypes for routine service provision.

Our group is developing enabling technologies that push the technical limits in spatial biology research and applying them for understanding how molecular and spatial composition of cells relates to their phenotype. We have set up cutting-edge methods that integrate advanced imaging and omics (SABER, Light-Seq) down to subcellular spatial omics, and implement automated workflows for hardware-software communication for adaptive feedback microscopy. We are particularly interested in spatial organization of subcellular compartments and their transcriptomic and proteomic composition. We have recently developed a new in situ proximity detection approach (ProPER) and would like to develop new applications for high-throughput in situ annotation of molecular states of RNA and proteins, high-fidelity detection of small and/or surface RNAs, as well as cell-cell interactions. We would also incorporate this approach together with other commercial and custom spatial omics methods.

Volume electron microscopy (vEM) is a fast-evolving imaging field that aims at acquiring ultrastructural details of cells, small model organisms, and tissues in three dimensions. vEM generally produces large datasets that can reach multiple TB of data, for which extracting valuable morphometric information by image segmentation is still a challenge. For targeted studies, which restrict measurement to selected regions of interest, other imaging modalities (light microscopy or X-ray microscopy) are used to build volumetric maps of the samples in which sub-regions of interest are defined. Due to the size and complexity of such multi-modal image datasets, it is challenging to efficiently (ideally automatically) extract biophysical information from these information-rich datasets. To tackle this challenge, the Bioimage Analysis Support Team and Electron Microscopy Core Facility from EMBL Heidelberg, as well as the VIB Bio Imaging Core from KU Leuven, are teaming up to develop and deploy reusable computational workflows for the automated management and AI-based analysis of such correlative volume EM datasets.

The sequence families team develops the InterPro, Pfam, Rfam and RNAcentral databases. This provides and excellent environment to learn and help develop cutting edge information resources. At present we are rapidly adopting AI technologies to assist in coding and curation. There are many opportunities and we are increasing our ambitions to provide detailed and accurate biological information more rapidly to the community. Projects could focus on using AI to improve data, or the software infrastructure for any of the data resources we provide.

Our team develops and applies Biological Small-Angle X-ray Scattering (BioSAXS) methods to support structural biology research, combining experimental development, beamline operation, and user service provision. We advance SAXS methodologies, including high-throughput and complex sample-environment approaches, while continuously improving beamline infrastructure, data workflows, and user-facing services. Current and future developments may involve both computational and experimental directions, such as data-processing pipelines, databases, automation strategies, or innovative sample environments. Fellows will have the opportunity to contribute according to their expertise, whether in computational method development, experimental instrumentation, sample-environment innovation, or integrated service provision within a collaborative and multidisciplinary beamline environment.

Our team focuses on developing and optimising operational pipelines for high‑quality biological sample generation, with a particular emphasis on early‑stage tissue handling, standardised processing workflows, and scalable cohort logistics. These pipelines form the foundation for downstream multi‑omics, imaging, and AI‑driven analyses, and are essential for ensuring reproducibility and FAIR data integration across projects. We are currently expanding our capabilities to support the Lifecourse programme, where efficient and standardised sample processing has been identified as a major bottleneck for future scaling. The group provides an ideal environment for fellows to learn workflow engineering, automation‑ready sample handling, quality control, and service‑oriented project management, while contributing to the development of robust technologies that support external users.

The EMBL Rome Light Imaging Facility is seeking ARISE2 postdoctoral fellows to develop automated imaging technology and image analysis for large-scale biological research. We develop technology that transforms service provision in imaging-based spatial omics as well as general areas of fluorescence microscopy and image analysis. Possible projects include “lab-in-the-loop” approaches for smart microscopy, robotics, and automation, as well as software development pipelines for image analysis.

Our group is developing scalable technologies to modernise expert biocuration, moving from predominantly manual workflows towards AI-assisted curation while maintaining UniProt’s high standards of quality, reliability, and scientific accuracy. Current and future work focuses on developing human-in-the-loop AI models and curation tools that can extract, structure, and validate biological knowledge from the literature and other data sources at scale. An ARISE2 fellow would contribute to the development of specialised AI models for vaccine capsular antigen curation, enabling relevant antigen information to be captured, standardised, and delivered through UniProt for use by the wider research community. This technology will support FAIR data management by converting unstructured biological knowledge into structured, traceable, interoperable, and reusable annotations, with community feedback used to refine the models and improve service provision for external researchers.

The Marquez Team has pioneered the development of Online Crystallography; fully automated protein-to-structure pipelines integrating crystallization, synchrotron data collection, and crystallographic data analysis into continuous workflows operated via the web. These pipelines rely on the CrystalDirect technology and the Crystallography Data Management system and have been fully integrated with automated data collection at the ESRF synchrotron, enabling rapid structure determination and large-scale X-ray-based fragment screening. Our facility generates large amounts of AI-ready data, and we are now extending our work to introduce Machine Learning (ML) approaches for AI-driven automation and for hit-to-lead optimisation and structure-based drug design. Our interdisciplinary team offers opportunities for scientists, engineers, and software developers to work in one of the leading infrastructures for structural biology within the areas of protein crystallography, fragment screening, drug design, automation, and large-scale scientific data management and Machine Learning. Currently, we are particularly interested in profiles in structural biology or computer science oriented towards one or several of the following areas: fragment screening, structure-guided drug design, machine learning, and artificial intelligence.

The Papp Team at EMBL Grenoble develops scientific instrumentation for macromolecular X-ray Crystallography and Cryo-Electron Microscopy applications. One of our most recent and significant development projects is the EasyGrid technology, a system designed to fully automate sample preparation and quality control for cryo-imaging experiments. Pilot measurements demonstrated that the EasyGrid system is suitable for preparing samples for Single Particle Analysis experiments and showed the superiority of EasyGrid-prepared samples in terms of ice quality for Cryo-Electron Tomography and X-ray Nano Imaging techniques, compared to traditional plunge freezing equipment. Based on these solid technical developments, the team is now working on creating software tools to fully automate the Cryo-EM/ET sample preparation workflow, including an AI-based system for preparation parameter optimization using an on-the-fly sample quality control method.

The Gene Editing and Virus Facility establishes ethical frameworks, designs strategies, develops reagents, optimises delivery, and ensures accurate detection of gene-edits. All these elements are areas of active innovation and evolution. For example, recently, we established high-frequency targeted DNA knock-in via viral delivery into mouse embryos, which was further developed using an all-in-one AAV containing an evolved novel mini-Cas9, gRNA, and repair template. Future work is to take this in-house expertise and expand the functionality of these tools across a more diverse range of organisms within the Kingdom Animalia.

Our group is developing enabling technologies that push the technical limits in spatial biology research and applying them for understanding how molecular and spatial composition of cells relates to their phenotype. We have set up cutting-edge methods that integrate advanced imaging and omics (SABER, Light-Seq) down to subcellular spatial omics, and implement automated workflows for hardware-software communication for adaptive feedback microscopy. We are particularly interested in spatial organization of subcellular compartments and their transcriptomic and proteomic composition. We have recently developed a new in situ proximity detection approach (ProPER) and would like to develop new applications for high-throughput in situ annotation of molecular states of RNA and proteins, high-fidelity detection of small and/or surface RNAs, as well as cell-cell interactions. We would also incorporate this approach together with other commercial and custom spatial omics methods.