Open Imputation Server provides secure Imputation services with provable genomic privacy

Abstract

As DNA sequencing data is available for personal use, genomic privacy is becoming a major challenge. Nevertheless, high-throughput genomic data analysis outsourcing is performed using pipelines that tend to overlook these challenges. Results: We present a client-server-based outsourcing framework for genotype imputation, an important step in genomic data analyses. Genotype data is encrypted by the client and encrypted data are used by the server that never observes the data in plain. Cloud-based framework can benefit from virtually unlimited computational resources while providing provable confidentiality. Availability: Server is publicly available at https://www.secureomics.org/OpenImpute. Users can anonymously test and use imputation server without registration.

Privacy-Aware Kinship Inference in Admixed Populations using Projection on Reference Panels

“…6b). As the data encryption is implemented in our protocol, even untrusted entities can be used in federated kinship estimation for making use of large cloud-based scaling for improved performance [65,66].…”

Privacy-preserving Polygenic Risk Scoring using Homomorphic Encryption

“…Blatt et al demonstrated that GWAS analysis of a real data set consisting of 25.000 individuals can be executed practically on encrypted data [4]. Recently, Harmanci et al developed a secure imputation web server based on homomorphic encryption where untyped variant data is predicted from available genotype data with the help of a reference panel [10]. The applications of HE is becoming more available and practical in genomics area.…”

“…Recently, new libraries and implementations such as Microsoft's SEAL and IBM's Fully Homomorphic Encryption Toolkit For Linux have been made available for developing HE applications. These tools make HE applications more practical and available for a variety of cases including genomics analysis [15,10,4].…”

Evaluation of vicinity-based hidden Markov models for genotype imputation

“…This way, the architecture of imputation algorithms on a cloud can be structured accordingly, for example, by using different models in different parts of the genome. On another front, the recent efforts to develop privacy-aware genotype imputation methods make use of the vicinity-based models to perform imputation while genotype data confidentiality is preserved [47,48]. These methods can potentially serve as alternative for genotype imputation servers such as Michigan Imputation Server [49], which provide imputation-as-a-service.…”

“…As the genetic privacy is becoming an important topic of public discussion, it is necessary to develop more efficient imputation methods that can be used to build the imputation services with privacy-by-design principles. While there have been attempts (including our group) to build secure imputation tools [47,48,50,51] using trusted execution methods and homomorphic encryption techniques [52], these methods are bound by computational requirements [53]. Therefore, our study can inform these methods about the locality parameters that must be considered and evaluated for decreasing computational requirements and maintaining the imputation accuracy while providing privacy and confidentiality for the genetic data.…”