Potential neutralizing antibodies discovered for novel corona virus using machine learning

Abstract The fast and untraceable virus mutations take lives of thousands people before the immune system can produce inhibitory antibody. recent outbreak COVID-19 infected killed in world. Rapid methods finding peptides or antibody sequences that inhibit viral epitopes SARS-CoV-2 will save life thousands. To predict neutralizing antibodies for a high-throughput manner, this paper, we use different machine learning (ML) model to possible synthetic SARS-CoV-2. We collected 1933 virus-antibody their clinical patient neutralization response trained an ML response. Using graph featurization with variety methods, like XGBoost, Random Forest, Multilayered Perceptron, Support Vector Machine Logistic Regression, screened hypothetical found nine stable potentially combined bioinformatics, structural biology, Molecular Dynamics (MD) simulations verify stability candidate