FEOpti-ACVP

FEOpti-ACVP predicts novel anti-coronavirus peptide sequences through feature engineering optimization combined with deep representation learning.

Key Features:

• Feature Engineering Optimization: The model employs optimized feature engineering as a core component for peptide sequence representation.
• Deep Representation Learning: The framework integrates deep representation learning for feature extraction from peptide sequences.
• Dual Framework Pre-training: The model was pre-trained using two distinct feature extraction frameworks.
• Multi-algorithm Evaluation: Several machine learning approaches were tested to construct the final prediction algorithm.

Scientific Applications:

• Anti-coronavirus Peptide Prediction: Identifies peptide sequences that inhibit adsorption and fusion of coronavirus with human cells.
• ACVP Drug Design: Supports computational screening of peptide-based inhibitors as potential therapeutic drug candidates.

Methodology:

The model combines feature engineering optimization with deep representation learning, was pre-trained using two feature extraction frameworks, and evaluated several machine learning approaches to construct the final algorithm.