The 2019 novel coronavirus infectious disease (COVID-19) pandemic has resulted in an unsustainable need for diagnostic tests. Currently, molecular tests are the accepted standard for the detection of SARS-CoV-2. Mass spectrometry (MS) enhanced by machine learning (ML) has recently been postulated to serve as a rapid, high-throughput, and low-cost alternative to molecular methods. Automated ML is a novel approach that could move mass spectrometry techniques beyond the confines of traditional laboratory settings. However, it remains unknown how different automated ML platforms perform for COVID-19 MS analysis. To this end, the goal of our study is to compare algorithms produced by two commercial automated ML platforms (Platforms A and B). Our study consisted of MS data derived from 361 subjects with molecular confirmation of COVID-19 status including SARS-CoV-2 variants. The top optimized ML model with respect to positive percent agreement (PPA) within Platforms A and B exhibited an accuracy of 94.9%, PPA of 100%, negative percent agreement (NPA) of 93%, and an accuracy of 91.8%, PPA of 100%, and NPA of 89%, respectively. These results illustrate the MS method's robustness against SARS-CoV-2 variants and highlight similarities and differences in automated ML platforms in producing optimal predictive algorithms for a given dataset.

Understanding disease burden and transmission dynamics in resource-limited, low-income countries like Nepal are often challenging due to inadequate surveillance systems. These issues are exacerbated by limited access to diagnostic and research facilities throughout the country. Nepal has one of the highest COVID-19 case rates (915 cases per 100,000 people) in South Asia, with densely-populated Kathmandu experiencing the highest number of cases. Swiftly identifying case clusters (hotspots) and introducing effective intervention programs is crucial to mounting an effective containment strategy. The rapid identification of circulating SARS-CoV-2 variants can also provide important information on viral evolution and epidemiology. Genomic-based environmental surveillance can help in the early detection of outbreaks before clinical cases are recognized and identify viral micro-diversity that can be used for designing real-time risk-based interventions. This research aimed to develop a genomic-based environmental surveillance system by detecting and characterizing SARS-CoV-2 in sewage samples of Kathmandu using portable next-generation DNA sequencing devices. Out of 22 sites in the Kathmandu Valley from June to August 2020, sewage samples from 16 (80%) sites had detectable SARS-CoV-2. A heatmap was created to visualize the presence of SARS-CoV-2 infection in the community based on viral load intensity and corresponding geospatial data. Further, 47 mutations were observed in the SARS-CoV-2 genome. Some detected mutations (n = 9, 22%) were novel at the time of data analysis and yet to be reported in the global database, with one indicating a frameshift deletion in the spike gene. SNP analysis revealed possibility of assessing circulating major/minor variant diversity on environmental samples based on key mutations. Our study demonstrated the feasibility of rapidly obtaining vital information on community transmission and disease dynamics of SARS-CoV-2 using genomic-based environmental surveillance.