WP6 - Privacy, Security and Trust
The novel data management techniques that will be developed in the project can see effective exploitation and social acceptance only if there is assurance that
sensitive information is not improperly leaked to unauthorized eyes. The objective of WP6 is the development of techniques for ensuring the respect of security and
privacy in the different scenarios where genomic data are stored, shared, processed, or disseminated.
TASK 6.1 CRYPTOGRAPHIC TECHNIQUES
Due to their sensitive nature, genomic data will be often maintained in encrypted form. We will develop innovative, effective, and efficient techniques for managing
and processing encrypted information. Existing solutions in this respect suffer from high complexity and are not suitable for genomic applications where computations
can involve terabytes of data. We will propose secure multi-party computation protocols to allow two parties holding genomic data to compute results based on both
data sets (e.g., finding a match) without disclosing the source data themselves.
TASK 6.2 QUERY PROCESSING AND PROVENANCE
Like in other fields, health and genomic applications will see an increasing reliance on external parties and cloud services for storing and processing data. We will
investigate solutions for ensuring protection of data when they are managed or processed by external parties, as well as for providing privacy of queries over the data
(which may leak information on the users submitting the queries). We will also enrich the genomic data model with provenance information and will allow reasoning
on it for data and process tracking for assessing data quality and trust.
TASK 6.3 DATA PROTECTION AND DISSEMINATION CONTROL
Genomic data could be abused, leading to discrimination (e.g., insurance companies using genetically-based diagnostic tests for determining premiums) or societal
stigmatization. we will design novel techniques for ensuring sensitive information is properly protected whenever genetic data need to be processed, published, or
shared. It will develop privacy techniques and metrics for evaluating the privacy ensured and the utility enjoyed (in terms of completeness and correctness of
supported queries) by the produced sanitized data. It will also provide access control solutions for supporting selective access to the data, as different parties (e.g.,
hospitals, government agencies, insurance companies, physicians, individuals) may be allowed for different views and actions over the data.
sensitive information is not improperly leaked to unauthorized eyes. The objective of WP6 is the development of techniques for ensuring the respect of security and
privacy in the different scenarios where genomic data are stored, shared, processed, or disseminated.
TASK 6.1 CRYPTOGRAPHIC TECHNIQUES
Due to their sensitive nature, genomic data will be often maintained in encrypted form. We will develop innovative, effective, and efficient techniques for managing
and processing encrypted information. Existing solutions in this respect suffer from high complexity and are not suitable for genomic applications where computations
can involve terabytes of data. We will propose secure multi-party computation protocols to allow two parties holding genomic data to compute results based on both
data sets (e.g., finding a match) without disclosing the source data themselves.
TASK 6.2 QUERY PROCESSING AND PROVENANCE
Like in other fields, health and genomic applications will see an increasing reliance on external parties and cloud services for storing and processing data. We will
investigate solutions for ensuring protection of data when they are managed or processed by external parties, as well as for providing privacy of queries over the data
(which may leak information on the users submitting the queries). We will also enrich the genomic data model with provenance information and will allow reasoning
on it for data and process tracking for assessing data quality and trust.
TASK 6.3 DATA PROTECTION AND DISSEMINATION CONTROL
Genomic data could be abused, leading to discrimination (e.g., insurance companies using genetically-based diagnostic tests for determining premiums) or societal
stigmatization. we will design novel techniques for ensuring sensitive information is properly protected whenever genetic data need to be processed, published, or
shared. It will develop privacy techniques and metrics for evaluating the privacy ensured and the utility enjoyed (in terms of completeness and correctness of
supported queries) by the produced sanitized data. It will also provide access control solutions for supporting selective access to the data, as different parties (e.g.,
hospitals, government agencies, insurance companies, physicians, individuals) may be allowed for different views and actions over the data.