Search

Since the discovery of coronavirus disease 2019 (COVID-19) in December 2019, it has been mainly diagnosed with quantitative reverse transcription polymerase chain reaction (PCR) of nasal swabs in clinics. A very sensitive and rapid detection technique using easily collected fluids such as saliva is needed for safer and more practical, precise mass testing. Here, we introduce a computationally screened gold-nanopatterned metasurface platform out of a pattern space of 2100 combinations for strongly enhanced light–virus interaction using a genetic algorithm and apply them to investigate the presence and concentration of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). In our approach, the gold metasurface with the nanopattern that provides the highest plasmonic enhancement is modified with the primary DNA aptamer for COVID-19 sensing from unprocessed saliva. A fluorescently tagged secondary aptamer was used to bind the virus that was then captured on the surface with the primary aptamer. By incorporating machine learning to identify the virus from Raman spectra, we achieved 95.2% sensitivity and specificity on 36 SARS-CoV-2 PCR-positive and 33 SARS-CoV-2 PCR-negative samples collected in the clinics. In addition, we demonstrated that our nanoplasmonic aptasensor could distinguish wild-type, Alpha, and Beta variants through the machine learning analysis of their spectra. Our results may help pave the way for effective, safe, and quantitative preventive screening and identification of variants.