A Computational Framework for Genomics

ABSTRACT. Genomic data comparison and storage play a critical role in forensic genetics
and large-scale genomic investigations, yet traditional approaches often require days to
process and compare complex genetic datasets. This paper presents an open-source,
CPU-based computational framework that reduces genomic comparison times from days
to seconds while maintaining forensic-grade accuracy. The proposed system builds upon a
kernel-based genomic representation combined with optimized data structures and parallel
processing strategies, enabling high-throughput analysis on classical hardware.
Unlike GPU-dependent solutions, the framework is designed to operate eciently on
standard personal computers, signicantly lowering hardware requirements and deploy-
ment costs. By adopting a fully local execution model, the proposed approach enhances
data privacy, security, and reproducibility, which are essential in forensic and institutional
genomic workows. The open-source implementation further promotes transparency and
methodological interpretability, facilitating adoption by domain experts. Performance
evaluations conducted on benchmark forensic DNA datasets demonstrate substantial im-
provements in processing speed while preserving the accuracy and reliability required
for forensic applications. Overall, this work establishes a scalable and accessible com-
putational foundation for rapid genomic comparison, with potential applicability beyond
forensic genetics to other large-scale genomic analysis domains.