AggreProt

AggreProt exploits an ensemble of deep neural networks to predict aggregation-prone regions in protein sequences and guide protein engineering strategies towards decreased aggregation propensity and improved solubility.

Key Features:

• Per-residue aggregation propensity prediction: Provides per-residue numerical aggregation propensity values and overall protein aggregation propensity profiles.
• Ensemble deep neural network architecture: Employs an ensemble of deep neural networks trained on experimentally evaluated hexapeptides.
• Solvent accessibility prediction: Reports the protein solvent accessibility area alongside aggregation predictions.
• Transmembrane propensity prediction: Provides transmembrane propensity information to aid in protein analysis.
• Interactive sequence and structure viewers: Enables comprehensive analysis through interactive visualization capabilities.

Scientific Applications:

• Aggregation-prone region identification: Identifies aggregation-prone regions essential for understanding misfolding diseases.
• Protein solubility engineering: Guides protein engineering strategies towards decreased aggregation propensity and improved solubility in recombinant proteins for industrial biocatalysts or therapeutics.
• Differential aggregation behavior prediction: Predicts differential aggregation behaviors in proteins across various use cases.

Methodology:

AggreProt uses an ensemble of deep neural networks trained on experimentally evaluated hexapeptides.