Robust Deep Neural Network-Based Framework for Predicting and Classifying Capsid Protein Based on Biomedical Data

Capsid protein is a pathogenic protein that needs to be examined because it helps in the virus's proliferation and mutation. Due to this protein, the virus can replicate and reproduce itself. The virus's outer boundary is made of capsid protein. Capsid protein analysis and prediction are essential. Several approaches, including mass spectrometry, have been developed to detect and predict Capsid protein. However, these methods are time-consuming and expensive and require highly skilled human resources. Therefore, this study proposed an efficient and robust classification approach for Capsid protein. The proposed model employs several machine learning, data science, and pattern recognition strategies to measure statistical moments based on obtained data. The experimental analysis reveals that the proposed model has achieved an overall 99% accuracy. These marks indicate that the suggested method outperformed the cutting-edge methods for classifying Capsid and non-Capsid proteins.