The main objective of GenData 2020 is the design of novel and advanced technological solutions for supporting the next-generation healthcare systems. We envision a society where genomic medicine will be completely integrated in patient care, and where genomic data will be part of the medical information associated with each of us. If on one hand this scenario has a positive impact on our lives (e.g., genomic diagnosis will allow each individual to be treated according to a personalized therapy), on the other hand it raises several issues, ranging from genomic data modeling and querying to the need of protecting privacy of sensitive information. In this project, UNIMI will focus on the privacy and security aspects that are key factors for the successful integration of genomics in healthcare: How do we ensure that our sensitive genomic data remain properly protected? How do we guarantee that the privacy of the individuals participating in medical studies is preserved? How do we guarantee that genomic data are accessed and used only by authorized parties? UNIMI will provide an answer to these and other questions by developing innovative solutions covering the different security and privacy issues related to the storage, collection, dissemination, and processing of genomic data.
UNIMI will lead WP6 on security, privacy and trust, and will participate in WP1 and WP7 on genomic data modeling and use cases, respectively. We now briefly describe the goal of these WPs and the main contributions of UNIMI in them.
WP1 will design the GenData 2020 data model, incorporating the different features of genomic information and providing a high-level query language for selective data extraction. UNIMI will collaborate to the definition of the GenData 2020 data model to guarantee that the model captures the concepts needed for specifying and enforcing privacy and security controls. UNIMI will also develop a formal rigorous semantics of (part of) the conceptual model, and will investigate the definition of a formal executable semantics for the query language.
WP6 will focus on technological solutions for ensuring the protection of genomic data and applications (e.g., genetic tests). Due to the highly sensitive nature of genomic data, their storage and processing will often be in encrypted form. UNIMI will collaborate to the identification of the properties that such cryptographic solutions for genomic data and applications should satisfy. UNIMI will also develop innovative data protection techniques that, by taking into consideration the peculiarities of genomic data, will ensure that private information will not be improperly leaked to unauthorized parties when data (or views of them) are published or shared, and that privacy and integrity be preserved even in scenarios where information may be externally stored or processed. UNIMI will develop an access control model, language and mechanism allowing the specification and enforcement of access authorizations on genomic data. The model and language will be expressive and flexible, and will consider different contextual information such as, for example, kind of users, roles, purpose of use, data provenance, or information sensitivity. Finally, UNIMI will develop metrics for assessing both the privacy risks associated with the publication and sharing of given genomic information and the utility enjoyed by sanitized data views.
WP7 will be devoted to the validation of the solutions developed in GenData 2020. UNIMI will be directly involved in the third use case dedicated to the production of anonymized and sanitized releases of data. In particular, UNIMI will contribute to the identification of the security and privacy requirements of the considered scenario and to the identification and enforcement of the techniques, developed in WP6, applicable to the use case.