SECURE CLOUD-BASED FRAMEWORK FOR GENOMIC DATA STORAGE AND ANALYSIS

Abstract

The genome files have sensitive biological, familial and disease related information and require a secure computational environment for data storage and analysis. This study proposed and tested a safe and secure cloud-based platform for storing and analyzing genomic data. The framework was intended to provide for encrypted storage of files, pseudonymization of metadata management, verification of integrity, approval of access control, secure execution of analysis, encrypted storage of results, controlled variant queries and audit logging. Using generated FASTQ and Variant Call Format datasets, a computational prototype was implemented in Python. They utilized de-identification procedures, integrity verification via checksums, authenticated encryption, role-based and attribute-based access control, consent-aware approval, and temporary workspaces for secure analysis in the framework. Both genomic files were successfully uploaded, encrypted, analyzed and logged by the prototype. All output from the FASTQ quality control analysis and the summary analysis of the Variant Call Format were generated in protected temporary workspaces and stored in encrypted format. The security evaluation found that there was no unauthorized upload by an auditor, no access for commercial purposes, and access for approved research was allowed. The results of the performance evaluation demonstrated low upload, encryption and analysis run-times in the simulated environment, which suggests that it is technically viable for small-scale genomic processing. Major workflow and security events were recorded in audit logs, aiding in traceability and accountability. The results demonstrate that two key aspects of a secure genomic cloud infrastructure should be encryption and consent governance, access control, workflow isolation, and controlled query disclosure. The suggested framework offers a repeatable model of storage and analysis for genomic data, while ensuring privacy. The framework should be validated with more genomic data, production cloud infrastructure, advanced key management, federated authentication, and more effective privacy-preserving analytical techniques in the future.