Teaming for Excellence  

The goal of the Teaming for Excellence initiative is to establish a world-class research and innovation hub in Kraków, Poland. The Centre will focus its activities on computational diagnostics, a field at the forefront of technological transformation in healthcare. By leveraging advanced statistical models, machine learning, and large-scale data analysis, it will play a pivotal role in the prevention, diagnosis, and treatment of complex diseases.

The Centre aims to streamline healthcare systems, reduce treatment costs, and improve patient outcomes across Europe. As a catalyst for change, it seeks to reshape the future of medicine through computational innovation.

NEARDATA 

The goal of  NEARDATA  is to create an extreme data infrastructure mediating data flows between Object Storage and Data Analytics platforms across the Compute Continuum. Novel XtremeDataHub platform is an intermediary data service that intercepts and optimises data flows (S3 API, stream APIs) with high performance near-data connectors (Cloud/Edge).   

ISW (In Silico World) 

The In Silico World project aims at accelerating the uptake of modelling and simulation technologies used for the development and regulatory assessment of medicines and medical devices, by lowering seven identified barriers: development, validation, accreditation, optimisation, exploitation, information, and training. 

This initiative employs computer models that leverage experimental data for hypothesis testing and outcome prediction. 

GEMINI

GEMINI is a project which promises to save lives and enhance the well-being of stroke patients, as it aims at improving diagnosis and treatment for acute ischaemic and haemorrhagic stroke.
This aim will be achieved by developing patient-specific decision-making tools and well-established models that can accurately assist in the diagnosis and stratification of stroke patients for tailored treatments.
Multi-organ and multi-scale Digital Twins models will be developed to improve our understanding of and support personalised treatment selection for this severe condition.

Weave 2023/05/Y/NZ2/00080  

Title: Evolutionary-scale interpretation of protein functions in the human gut microbiome. 

The Weave project aims to create a detailed atlas of human gut protein structures with functional annotations, centered around a Protein Universe Map for navigating the gut microbiome. Leveraging deep learning and evolutionary modeling, it focuses on the 40% of gut proteins with unknown functions. By using advanced tools like AlphaFold2 and deepFRI, overcomes the limitations of traditional annotation methods and paves the way for deeper insights into gut microbiota metabolism and future therapeutic discovery. 

Project Goals: 

• Catalogue Construction: Assemble a detailed library of gut protein structures and their predicted functions, including modeling novel 3D structures. 
• Diversity Exploration: Analyze the sequence and structural diversity of functionally unknown proteins within the global protein space. 
• Evolutionary Tracing: Investigate evolutionary origins and potential relationships of gut proteins across different ecosystems. 
• Function Prediction: Combine structural data with genomic context to infer biological functions of dark proteins. 

Duration: 19 January 2024 – 18 January 2028 

Project lead: Tomasz Kościółek (Research Team Leader of Structural and Functional Genomics Group) 

Partners/ List of collaborators: 

• University of Basel

Preludium 2023/49/N/ST6/04252

Title: Understanding the influence of variation in venous anatomy on local haemodynamics in patients with deep vein thrombosis of the lower limb: statistical shape modelling approach. 

This project aims to improve the understanding and prediction of post-thrombotic syndrome (PTS) following deep vein thrombosis (DVT) in the lower limbs by investigating how local venous anatomy influences blood flow patterns associated with thrombus formation. Using a combination of Computational Fluid Dynamics (CFD) and Statistical Shape Models (SSM), the research will analyze both idealized and patient-specific venous geometries derived from medical imaging (CT/MRI) to identify key anatomical features linked to abnormal haemodynamics, such as low wall shear stress and retrograde flow. 

The core objective is to determine whether SSMs can efficiently characterize shape-related risk factors and reduce the need for computationally expensive CFD simulations in clinical workflows. By applying sensitivity analysis and uncertainty quantification methods, the project will establish a framework to assess how anatomical variability influences haemodynamic metrics, ultimately supporting more accurate and cost-effective patient-specific diagnosis and intervention planning. This novel application of SSM-CFD integration to venous disease has strong potential to enhance clinical decision-making in vascular medicine. 

Project Goals:  

The influence of the diversity of vein anatomy on the local anatomy hemodynamics in patients with venous thrombosis  deep lower limb: statistical technician shape modeling. 

Duration:  29 January 2024 – 28 August 2025 

Project lead: Magdalena Otta (PhD Student in Extreme-scale Data and Computing)

Partners/ List of collaborators: 

• University of Sheffield 
• Royal Free Hospital 

Preludium  2024/53/N/NZ4/03513 

Title: Diffusion magnetic resonance imaging for accurate white matter tracking in peritumoral tissue to enhance surgical planning, precision, and prediction in neuro-oncology  

This project leverages diffusion MRI (dMRI) and tractography to map glioma-induced structural and connectivity changes in the brain. By combining imaging data with molecular markers, and applying advanced techniques such as spherical deconvolution, the project aims to improve fiber tracking in tumor-adjacent regions and provide deeper insights into tumor behavior. The ultimate goal is to develop reliable imaging biomarkers to support diagnosis, surgical planning, and understanding of glioma biology. 

Project Goals: 

• Optimize preprocessing pipelines for large-scale dMRI datasets of glioma patients. 
• Apply advanced tractography techniques, including spherical deconvolution, to improve reconstruction of white matter in peritumoral regions. 
• Integrate imaging and molecular data to explore the relationship between tumor biology and structural brain connectivity. 
• Support the development of imaging biomarkers to enhance glioma diagnosis, treatment planning, and research, especially when individual dMRI scans are not available. 

Duration:  16 January 2025 – 15 January 2027

Project lead: Joan Falco Roget (PhD Student in Computational Neuroscience)

Partners/ List of collaborators: 

• The University of Messina