Detection and Identification of SARS-CoV-2 Variants Using Single-Molecule Conductance Measurements

Since the start of the 2019 pandemic (COVID-19), the development of rapid and cost-effective testing has been an ongoing effort to help combat this highly transmittable and infectious disease. Over the past two years, the virus has continued to mutate and evolve, and several variants of SARS-CoV-2 have emerged which have reduced the vaccines’ effectiveness. Therefore, there is a significant need for surveillance of the SARS-CoV-2 variants of concern to track the spread of the disease and help to confine the virus. Currently, the standard identification method for the subvariants utilizes reverse-transcription polymerase chain reaction (RT-PCR). However, this method is not suitable for rapid, high throughput testing of large populations, it is costly, time-consuming, and requires expert staff. Here we present an alternative to RT-PCR based on single-molecule conductance measurements which open new avenues for sensitive and cost-effective detection of the SARS-CoV-2 variants of concern.<br/>Previous studies using the single-molecule break junction (SMBJ) approach have demonstrated that this technique is capable of detecting and identifying RNA: DNA hybrids via their conductance values. This work focuses on identifying SARS-CoV-2 variants of concern based on the codon sequences that represent mutations in the spike protein. Our hypothesis is that using a specific DNA probe that is complementary to the RNA for every mutation would result in a specific conductance value that represents the presence of the variant in the sample. Therefore, we can detect and identify each variant in a mixed sample based on their unique electronic fingerprints.<br/>We focus on the alpha, beta, and delta variants which according to the World Health Organization (WHO) are classified as SARS-CoV-2 variants of concern. We design a series of 12 base-pair DNA probes that contain the mutation and would bind to the corresponding RNA section of the alpha and beta variant genome block. Our experimental results indicate that in the presence of the mutation, the conductance value of the hybrid RNA: DNA sequence is significantly different from when there is no mutation present. We further utilize the XGBoost machine learning classifier to aid in the rapid detection of the SARS-CoV-2 variants. This work demonstrates the utility of the SMBJ technique for the accurate detection and identification of SARS-CoV-2 variants with the potential for real-time application.