TPCP (Turin Prostate Cancer Prognostication): “An integrated approach for improving the prognosis of prostate cancer” is an observational non-pharmacological retrospective cohort study whose objective is to address the need to improve the prognosis of prostate cancer, both in terms of identification of new prognostic markers and in terms of improvement of current prognostic models.
The Turin Prostate Cancer Prognostication (TPCP) is a study promoted by the Department of Medical Sciences of the University of Turin, funded by the AIRC Foundation for Cancer Research and launched in 2021 at the AOU Città della Salute e della Scienza of Turin.

Contacts:

Contacts: Lorenzo Richiardi
web site: studio TPCP

Modern scientific software requires ever-increasing computing capacity, necessitating increasingly specialized systems to run on purpose-built supercomputers. Given that each supercomputer is generally a one-off project, the need for computing frameworks portable across different CPU and GPU architectures without performance losses is increasingly compelling. Pennycook’s portability score allows for quantifying how performance varies across HW platforms. We use it to study the properties of different HPC programs, for example, the solver module of the AVU-GSR pipeline for the ESA Gaia mission. This code finds the astrometric parameters of 108 stars in the Milky Way using the LSQR iterative algorithm. Besides its astrophysical relevance, LSQR is widely used to solve linear systems of equations across a wide range of high-performance computing applications. We optimized the previous CUDA implementation and ported the code to further six GPU-acceleration frameworks: C++ PSTL, SYCL, OpenMP, HIP, KOKKOS, and OpenACC. Thanks to the great availability of heterogeneous resources on HPC4AI, we evaluated each framework’s performance portability in terms of application and architectural efficiency across different GPUs. Specifically, we use NVIDIA T4, V100, A100, and the new GH200. For each GPU, we compute the roofline model to get the architectural efficiency.

Contacts:
Contact: Giulio Malenza

Writing Uses: Transmission of Knowledge, Administrative Practices and Political Control in Anatolian and Syro-Anatolian Polities in the 2nd and 1st Millennium BCE. Networks of Power: Institutional Hierarchies and State Management in Late Bronze Age Western Asia (NePo). Prof. Stefano De Martino (stefano.demartino@unito.it)

Contacts:

Contacts: Rossana Damiano

web site: https://hfpo.di.unito.it/

The collaboration with the HPC4AI project arises from the increasingly complex calculation needs of our faculty. The service allows us to add considerable computational power to the research activities of the Collegio Carlo Alberto.

Contacts:

Contacts: Alberto Galvan

web site:https://www.carloalberto.org/

Microorganisms and microbiomes are crucial for several research areas and applications (e.g. agri-food, environmental, biotechnological, biomedical sectors). Indeed, as whole organisms or as products (enzymes, antibiotics, etc), they can be harnessed for the production of foods, pharmaceuticals, and fragrances, for agricultural processes, and in fighting climate change. The Microbial Resource Research Infrastructure (MIRRI) is the pan-European infrastructure in charge of the preservation, analysis, supply, and valorization of microbial resources and biodiversity. MIRRI has recently achieved the “Landmark” status on the European Strategy Forum on the ESFRI Roadmap 2021 and is working towards the European Research Infrastructure Consortium (ERIC) status.

Contacts:

Contacts: Marco Beccuti

web site: https://www.mirri-it.it/

Nowadays deep learning models can solve many challenging tasks like machine translation, image captioning, or text understanding, but results are still not completely satisfying and new ideas are needed to make further steps toward full Natural Language Processing. In this research area, modeling dialog is surely one of the most challenging application; systems need “to understand” arguments and dialog history, and to generate coherent and aligned text. The research activity will aim at studying new ways to leverage deep learning and deep reinforcement learning techniques to improve on dialog generation obtaining more precise and sensible outputs. The starting point will be to solve the question about how to include into these pipelines external source of information like knowledge bases through specialized retrieval systems, increasing model performance, reliability, and generation accuracy. Specialized libraries like Pytorch will be used to implement and prove such new approaches.


Contacts:
Rossella Cancelliere – rossella.cancelliere-at-unito.it
Giovanni Bonetta – giovanni.bonetta-at-unito.it

While it is clear that bacteria in the gut and neurons communicate (the Machine learning and deep learning-based approaches have emerged as key components in biomedical data analyses. Deep learning tools enhance the performance in pattern-recognition tasks. Their ability to learn complex patterns in data has tremendous implications in immunogenomics.
Our project focuses to the identification of disease specific T-cell receptor motifs. Specifically, we use and develop deep learning algorithms focusing on the extraction of antigen-specific TCR motifs from bulk and single-cell RNA-Seq.


Contacts:
Raffaele Calogero

While it is clear that bacteria in the gut and neurons communicate (the microbiota-gut- brain axis), regulating brain health and behavior, fundamental questions in this field are still wide-open. Most of the research carried out so far has focused on molecular mechanisms and not – as we propose – on top- down, information-based approaches, which are required to extract high-level meaning from biochemical mechanisms. Here, we will move beyond our initial Brain-bacteria Interface (BBI), the first integrated electrical-optical neural-bacteria to extract real-time information across biological entities, to develop the first hi-BBI. Taking advantage of the multidisciplinary and world-renowned expertise of the researchers which will be involved in the project, we will develop and exploit the hi-BBI and by coupling a complex artificial multi-organism platform with machine learning and innovative sensing devices, we aim to discover fundamental aspects of the bidirectional communication between gut bacteria and neurons.

Organisations involved: Complutense University of Madrid (UCM; Faculty of Biological Sciences; Madrid, Spain), University of Torino (Unito; Molecular Biotechnology Center; Torino, Italy), Politecnico di Torino (Torino, Italy), Tufts University (Boston, MA, US)

Contacts:
Stefano Geuna

The B-CRATOS project aims at creating the first battery-free high-speed wireless in-body communication platform for Brain-Machine-Body connectivity, to control a robotic hand prothesis. HPC and embedded AI models will play an essential role in the system since the complex deep learning models needed to translate neural signals into meaningful commands for the robotic hand will be trained by LINKS Foundation on high-performance, GPU-equipped computing infrastructure. These models will also require a continual learning approach to compensate for the time variability of the neural signals. Finally, these models will be distilled and deployed on the very low-power devices used to control the prothesis.


Contacts:
Paolo Viviani (paolo.viviani@linksfoundation.com)
SiteWeb: http://www.b-cratos.eu/