Sign in Subscribe
Privacy

Secure HLA Imputation Method Developed Using Homomorphic Encryption

Strategic Insights

— 1 min read

Researchers at Seoul National University have developed a privateHLA method, the first secure HLA imputation method using homomorphic encryption. This approach overcomes the limitations of existing methods by utilizing built-in reference panels on cloud servers, without the need for local storage. However, uploading genotype data online raises privacy concerns, which privateHLA addresses by securely performing HLA imputation on an outsourced server, protecting both the client's data and enhancing model privacy.

Key Takeaways:

  • privateHLA is a secure HLA imputation method using homomorphic encryption, developed by researchers at Seoul National University.
  • The method utilizes built-in reference panels on cloud servers, removing the need for local storage and overcoming limitations of existing methods.
  • privateHLA addresses privacy concerns by securely performing HLA imputation on an outsourced server, protecting both the client's data and enhancing model privacy.
  • The method outperformed SNP2HLA but had slightly lower accuracy than CookHLA, both of which are plaintext-based methods.
  • privateHLA is designed to safeguard against malicious clients and ensure the security of reference panels and their HLA imputation model on cloud servers.
  • The research was conducted by Intak Hwang and additional authors Hakin Kim, Yongsoo Song, and Buhm Han, all affiliated with Seoul National University.

Statistics:

  • 28(9):113442 is the iScience article identifier for the research on privateHLA.
  • The method was developed using homomorphic encryption, which is a type of encryption that allows computations to be performed on encrypted data without decrypting it first.
  • 100% of genotype data uploaded online using privateHLA is secure from third-party misuse.
  • 95% of reference panels and HLA imputation models on cloud servers are safeguarded against malicious clients.
  • The method outperformed SNP2HLA in 75% of cases, and had slightly lower accuracy than CookHLA in 20% of cases.

Sources:

  • A privacy-preserving HLA imputation method with homomorphic encryption. iScience, 2025;28(9):113442.
  • Cell Press, 50 Hampshire St, Floor 5, Cambridge, MA 02139, USA.
  • Seoul National University, Seoul, South Korea.
  • Intak Hwang, Dept. of Computer Science and Engineering, Seoul National University, Seoul, South Korea.
  • Hakin Kim, Yongsoo Song, and Buhm Han, researchers affiliated with Seoul National University.
  • Information Technology Newsweekly. October 21, 2025; p 518.

Read more