Predicting Novel Coronavirus (nCoV) strains detecting the mutation process applying neural networking

Abstract

As viruses undergo rapid evolution, the SARS-CoV-2 which is known as Covid- 19 has persisted in human populations for approximately three and a half years rapidly, continually exhibiting swift and unpredictable mutations. The relentless emergence of various new strains of SARS-CoV-2 has posed a significant challenge, leaving researchers grappling for effective strategies. This study employs a machine learning approach known as the Seq2Seq model to predict future new variants of the Human Coronavirus family by using the genome sequences of Human Coronaviruses in time series manner based on their first evolution. Through this methodology, the research successfully predicts and generates the future possible variants genome sequence of Human Coronavirus. This model would be a useful tool to predict genome sequences of future Human Coronaviruses and get important insights of the future variants to tackle the problem of fast evaluation of the human coronaviruses.