PLoS ONE: Scientists use machine learning technology to successfully track the genome characteristics of COVID-19

, , May 12, 2020 /
PRNewswire2020 (UPI) -- In a study published in the international journal PLoS ONE , scientists from the University of Western Ontario and others used machine learning techniques to identify the potential genomic characteristics of 29 different COVID-19 DNA sequences The new data discovery tool can help researchers quickly and easily classify lethal viruses such as SARS-CoV-2 in minutes Picture Source: PLOS ONE (2020) DOI: 10.1371/journal.pone.0232391 This study supports the researchers' hypothesis that SARS-CoV-2 may have originated in the sarbecoviru subspecies of bats, which is a subgroup of the Betacoronaviru virus, the researchers said This ultra-fast, scalable, and highly accurate classification cell can be classified using a special graph-based software and decision tree approach, and can filter out the best selection for all possible results, using a graph-based specialized software to illustrate the best results in all detection possibilities The researchers say the machine learning method can achieve 100% accurate classification of COVID-19 sequences and, more importantly, reveal important correlations between more than 5,000 virus genomes in minutes; The classification tool has now been able to analyze more than 5,000 special viral genome sequences, including 29 COVID-19 sequences updated on January 27, and researcher Hill believes the new tool will not only classify any newly discovered COVID-19 virus sequences, but also serve as an essential component in the development of vaccines and new drug kits, and be used by front-line health care workers and scientists during the global pandemic (BiovalleyBioon.com) Original Origin: Gurjit S Randhawa, Maximillian P M Soltyiak, Hadi El Roz, et al.
Machine ing intrinic genomic ignature for clification of novel: COVID-19 cae tudy , PLOS ONE (2020) DOI: 10.1371/journal.pone.0232391