Enhancing public health outcomes with AI-powered clinical surveillance: Precise detection of COVID-19 variants using qPCR and nanopore sequencing

Background: We aimed to evaluate the efficacy of integrating the Varia5 multiplex assay (qPCR) and whole genome sequencing (WGS) for monitoring SARS-CoV-2, focusing on their overall performance in identifying various virus variants. Methods: This study included 140 naso-pharyngeal swab samples from individuals with suspected COVID19. We utilized our self-developed Varia5 multiplex assay, which targets five viral genes linked to COVID-19 mutations, in conjunction with comprehensive genomic analysis performed through whole genome sequencing (WGS) using the Oxford Nanopore system. Machine learning was integrated to optimize the qPCR conditions and enhance the detection efficiency. Results: The Varia5 assay identified the prevalent BA.2.75 variant in 92 samples compared to that in 81 samples detected via WGS. The BA.5.2 variant, indicative of higher viral loads, was identified in 15 samples via Varia5 and in 14 samples via WGS.Furthermore, rare variants, such as BA.2.10, were identified. The mean Ct value was 18.36, with significant viral load differences noted between specific variants. Conclusion: Our findings demonstrate that while WGS offers enhanced sensitivity and specificity for variant detection, qPCR remains crucial for large-scale testing because of its cost and time efficiency. The integrated approach, which combines both techniques, represents a more comprehensive monitoring algorithm that can improve public health strategies against pandemics such as COVID-19. (c) 2025 The Author(s). Published by Elsevier Ltd on behalf of King Saud Bin Abdulaziz University for Health Sciences. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).