Ultra-sensitive detection of receptor binding domain of SARS-CoV-2 based on artificial intelligence assisted antibody selection and antibody‒nucleic acid conjugation signal amplification

Abstract

The emergence of SARS-CoV-2 has led to a global pandemic, causing unprecedented disruption to health systems and economies worldwide, calling for an urgent imperative for precise and sensitive diagnostic method. To address this, a novel detection method, termed antibody‒nucleic acid conjugate linked immunosorbent assay (ANCLISA), was designed for the ultra-sensitive detection of the receptor binding domain (RBD) of SARS-CoV-2, which play a pivotal role in the diagnosis of COVID-19. Leveraging the Alphafold 2 model for simulating antibody‒antigen complexes and the ΔΔG predictor model for forecasting the binding free energy of antigen‒antibody complex, we successfully identified the high-affinity antibody TB1. Through the utilization of succinimidyl 4-[N-maleimidomethyl] cyclohexane-1-carboxylate (SMCC) for antibody‒nucleic acid conjugation and the integration of the hybridization chain reaction (HCR) for signal amplification within the system, we have established an ultra-sensitive approach for the SARS-CoV-2 Delta variant detection. The proposed immunoassay provided exhibited a linear detection range spanning 11 fg/mL to 11 ng/mL, achieving a limit of detection (LOD) of 1.48 fg/mL. Our innovative approach represents a significant stride in enabling precise and ultra-sensitive detection of SARS-CoV-2, thereby holding promise for enhancing diagnostic capabilities in the ongoing global battle against the pandemic.