Cascade Open Call 2024 for Research Node 1 - THE GUARDIAN OF THE INVISIBLE: ARTIFICIAL INTELLIGENCE AND EMERGING MICROORGANISMSn

Artificial intelligence (AI) has now become a no-longer-questionable tool in the study and tracking of emerging microorganisms. One of the main AI applications involves predictive models to analyse demographic, environmental, health, and travel data to predict the spread of emerging microorganisms. AI can be used to trace and reconstruct the contact network of infected patients in order to identify microorganisms' patterns and geographic distribution as well as critical points to implement control measures. However, AI possesses yet unexplored capability involving the combined use of genetic data, deep learning algorithms and advanced computational capabilities. Moreover, other future perspectives concern revealing both complex viral evolutionary dynamics and reservoir-pathogen-host interaction mechanisms aiming to assess potential risks associated with spillover events that could lead to future pandemics.

The project involves the development of several AI models (e.g., deep learning algorithms, support vector machines, neural networks, and decision trees) to analyse cross-sectional omics data in order to characterise critical variables (e.g., genetic markers) in the dynamics of viral evolutions. Consequently, further models will be necessary to search critical variables related to microorganism-host molecular interactions, in addition to possible spillover events (i.e., species jumping). AI models can provide valuable insights into the risk assessment of these events exploiting the deep understanding of genetic, environmental, and zoonotic factors. Critical variables' selection will be performed with advanced techniques in order to detect those that optimise predictive model accuracy. In addition, the ability of AI to analyse big data in real time is crucial for monitoring microorganism populations. Indeed, the continuous tracking and the analysis of fitness parameters would allow AI to promptly identify any deviation from basic models.

Expected Results:
Advanced AI algorithms (e.g., Markov chains, multi-task learning) will be used to predict the viral fitness landscape and explore antigenic evolution considering factors such as replication and recombination rates. Furthermore, those methods will be exploited in outbreak simulations before their emergence using the information available uniquely at the beginning of an outbreak to reveal new variants and/or new recombinant pathogens. AI techniques can be used to develop classification models, such as convolutional neural networks (CNNs), that independently classify pathogens into preset categories based on patterns and motifs found within genetic sequences. It is possible to predict the behavior of these functional motifs within the microorganism's proteins and calculate their impact on the overall proteins stability and three-dimensional structure, using molecular dynamics simulations and structure modeling . Concerning pathogen-host interaction, models based on unsupervised learning can be used to predict the binding specificity of a cellular receptor. However, predicting antibody binding specificity is much more complex and less explored in the literature. This approach can simplify preventive actions and the design of antibodies whose targets are minority variants intercepted before they become prevalent, as well as their use as reagents in routine diagnostics. Finally, AI models can be boost to create pathogenicity prediction mathematical models using patient-specific biological data (e.g., immunocompetent or immunocompromised status) to study targeted therapies for precision medicine.

• Calls will be open to teams including both private and public entities located in Sourthern Italy(minimum: 3, maximum: 5 entities), external to the INF-ACT consortium and to members of the beneficiary team of the COC-1-2023-UNIPV call.
• Each team must identify a lead organization (Leader), or the subject appointed and authorized to represent the whole team and maintain communications with the Spoke during all phases of the presentation and evaluation procedure and for the entire duration of the activities planning;
• The team Leader must be a public University or a public Research Organization;
• Each team Leader may submit only one project proposal in response to this call. On the other hand, the possibility is not precluded for the same entity to be part of other teams in response to other Cascade Calls
• Each team entity must indicate the name of the researchers involved in the proposed project activities (so-called "critical mass"), the expected commitment in person-months, their curriculum vitae to assess the actual contribution to the research work.
• Each team entity must participate with at least one researcher. The maximum number of researchers to be involved as critical mass for this project is 12.
• Entitites who are Team Leader in projects awarded by the COC-1-2023 are eligible only as beneficiaries;
• 100% of the requested financing must be dedicated to entities located in regions of Southern Italy (tagging: territorial);
• In case of recruitment of new researchers, the team must comply with the provisions of article 47 "Equal opportunities, generational and gender, in public contracts PNRR and PNC" of Law Decree 31-May-2021, no. 77 converted into law 29-July-2021, n. 108;
• Budget breakdown and financial reporting must comply with guidelines and regulation for RRNP funded-projects.

The call text (in Italian) has been published on the UNIPV website.

Collection of applications started at 12:00 PM on May 24, 2024 and closed at 12:00 PM on June 24, 2024
Evaluations are currently in progress

• All allocated funds must be spent by the end of October 2025. Should MUR allow time extensions to the NRRP/PNRR INF-ACT project, extensions might be also provided to COC projects;
• Awarded teams will comply to NRRP/PNRR reporting rules as per the indications that will be made by INF-ACT Spoke and INF-ACT Foundation;
• Awarded teams will have to join the INF-ACT foundation as partners;
• Funding should be acknowledged in all publications emanating from the research carried out during the award period together with NRRP/PNRR funds using the following statements: "This research was supported by EU funding within the NextGenerationEU-MUR PNRR Extended Partnership initiative on Emerging Infectious Diseases (Project no. PE00000007, INF-ACT). [BENEFICIARY NAME] was recipient of INF-ACT Cascade Open Call 2024 (ID CALL)."
• Participation of private entity and co-financing are not mandatory, but an added value;
• Financial capacity of private entity will be assessed ex-post by third parties. In case of negative outcome, the private entity will be excluded from participating to the awarded project and, therefore, from receiving funding.

Proposals are being evaluated on the basis of the following evaluation criteria:

• Objective and scientific quality (max 50 points)
• Innovation and private sector involvement (max 20 points)
• Impact (max 15 points)
• Economic sustainability (max 15 points)

Proposals will be considered admissible for financing if they obtain at least 70 out of 100 points. Evaluation will be carried out by a panel of 3-5 international and highly-qualified scientists, who are not directly involved in the INF-ACT Research Program. Reviewers will sign a declaration to exclude possible conflicts of interest before accessing the scientific proposals to be evaluated.