Safe genetics: INESC TEC researchers innovate cloud-based data processing

A team of researchers at INESC TEC has developed a new tool called Gyosa, capable of conducting genomic studies securely in cloud computing environments without compromising data privacy. 

This innovation is set to transform the way large-scale genomic analyses are performed, particularly in Genomic Wide-Association Studies (GWAS). The increasing volume of human genomic data requires the sharing and processing of vast datasets, often on external servers. The challenge? Ensuring that this data is handled securely in the cloud. 

“With this advancement, we are contributing to safer, more accessible, and more scalable genetic research, paving the way for new international collaborations and large-scale analyses – without compromising data confidentiality,” explained Cláudia Brito, INESC TEC researcher in storage systems and Artificial Intelligence (AI). 

To address this issue, Gyosa leverages trusted execution environments, such as Intel SGX, ensuring that data remains protected even when processed on third-party infrastructures. 

Beyond security, the system stands out in terms of efficiency and scalability in genomic data processing through distributed computing. Developed in collaboration with João Paulo, also a researcher at INESC TEC in storage systems, the solution has proven capable of processing up to 80,000 genomic files while maintaining processing accuracy. 

“Gyosa creates an environment to which data is securely transferred with encryption from one server to a remote server. The workload is distributed across multiple servers in virtual machines, making the analysis process more efficient,” explained Pedro Ferreira, INESC TEC researcher in AI and bioinformatics, who also contributed to the study. 

Based on the paper Exploiting Trusted Execution Environments and Distributed Computation for Genomic Association Tests, published in the leading journal IEEE Journal of Biomedical and Health Informatics, the name Gyosa was inspired by the Asian dumpling – symbolising the idea of gathering and processing data in a secure environment. 

The researchers mentioned in this news piece are associated with INESC TEC, U.Minho and UP-FCUP.