Effective of Various Vaccines on Antibody Response and Genetic Immune Using Deep Learning Method

Abstract

The COVID-19 epidemic has affected daily life on a global scale. Many research teams from
major pharmaceutical companies and university institutions around the world have been
developing vaccines since the beginning of the pandemic. The effectiveness, acceptability, and
results of vaccinations are influenced by gender. The SARS-CoV-2 mRNA vaccines were
released on the market in reaction to the Covid-19 public health emergencies. There is no history
of using mRNA vaccines to treat infectious diseases. The numerous modifications of the
vaccine’s mRNA work to protect it from cellular defenses, lengthen its biological half-life and
increase the creation of spike protein. In this paper, we propose a novel model to predict the
antibody response based on deep learning by proposing a Convolutional Neural Network (CNN)
model. The proposed system consists of several stages, where the GSE201533 dataset which
used, containing 26,370 features is first split into two sets, then the missing value and
normalization were applied as a preprocessing stage, then the best features were selected using
three techniques (Mutual information, Chi-square, and Analysis of Variance (ANOVA)). Then
the selected features were classified using the proposed CNN model. The proposed CNN
contributed to raising the accuracy of the model and reducing the time required for prediction.
The experimental results indicate an accuracy rate of 100% in all cases.