GAP

GAP classifies hexapeptides as amyloid fibril-forming or amorphous β-aggregating to inform biomaterial design and aggregation inhibitor development.

Key Features:

• High Accuracy: Achieves nearly 100% accuracy in validation tests using non-redundant datasets.
• Machine Learning Integration: Converts observed preferences of adjacent and alternate-position residue pairs in hexapeptides into energy potentials that serve as input features for machine learning classifiers.
• Focus on Amino Acid Side Chains: Emphasizes the role of amino acid side chains in determining the morphological outcomes of β-mediated aggregates formed by short peptides.

Scientific Applications:

• Biomaterial Design: Identifies amyloid fibril-forming versus amorphous β-aggregating hexapeptides to inform rational design of biomaterials with specified aggregation morphology.
• Therapeutic Development: Supports development of aggregation inhibitors for protein-misfolding diseases such as Alzheimer’s disease by predicting aggregation propensity of hexapeptides.

Methodology:

Translates distinct preferences of residue pairs in hexapeptides into energy potentials and uses those potentials as input features for machine learning algorithms that classify peptides by aggregation propensity.