RESUMEN
The outbreak of COVID-19 has spread worldwide, causing great damage to the global economy. Raman spectroscopy is expected to become a rapid and accurate method for the detection of coronavirus. A classification method of coronavirus spike proteins by Raman spectroscopy based on deep learning was implemented. A Raman spectra dataset of the spike proteins of five coronaviruses (including MERS-CoV, SARS-CoV, SARS-CoV-2, HCoVHKU1, and HCoV-OC43) was generated to establish the neural network model for classification. Even for rapidly acquired spectra with a low signal-to-noise ratio, the average accuracy exceeded 97%. An interpretive analysis of the classification results of the neural network was performed, which indicated that the differences in spectral characteristics captured by the neural network were consistent with the experimental analysis. The interpretative analysis method provided a valuable reference for identifying complex Raman spectra using deep-learning techniques. Our approach exhibited the potential to be applied in clinical practice to identify COVID-19 and other coronaviruses, and it can also be applied to other identification problems such as the identification of viruses or chemical agents, as well as in industrial areas such as oil and gas exploration.
RESUMEN
The outbreak of COVID-19 has spread worldwide, causing great damage to the global economy. Raman spectroscopy is expected to become a rapid and accurate method for the detection of coronavirus. A classification method of coronavirus spike proteins by Raman spectroscopy based on deep learning was implemented. A Raman spectra dataset of the spike proteins of five coronaviruses (including MERS-CoV, SARS-CoV, SARS-CoV-2, HCoVHKU1, and HCoV-OC43) was generated to establish the neural network model for classification. Even for rapidly acquired spectra with a low signal-to-noise ratio, the average accuracy exceeded 97%. An interpretive analysis of the classification results of the neural network was performed, which indicated that the differences in spectral characteristics captured by the neural network were consistent with the experimental analysis. The interpretative analysis method provided a valuable reference for identifying complex Raman spectra using deep-learning techniques. Our approach exhibited the potential to be applied in clinical practice to identify COVID-19 and other coronaviruses, and it can also be applied to other identification problems such as the identification of viruses or chemical agents, as well as in industrial areas such as oil and gas exploration.
RESUMEN
In order to control COVID-19, rapid and accurate detection of the pathogenic, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is an urgent task. The target spike proteins of SARS-CoV-2 have been detected experimentally via Raman spectroscopy. However, there lacks high-accuracy theoretical Raman spectra of the spike proteins to as a standard reference for the clinic diagnostic purpose. In this paper, we propose a large fragment method to construct the high-precision Raman spectra for the spike proteins. The large fragment method not only reduces the calculation error but also improves the accuracy of the protein Raman spectra by completely calculating the interactions within the large fragment. The Pearson correlation coefficient of theoretical Raman spectra is greater than 0.929 or more. Compared with the experimental spectra, the characteristic patterns are easily visible. This work provides a detection standard for the spike proteins which shall bring a step closer to the fast recognition of SARS-CoV-2 via Raman spectroscopy method.