Tango

Tango predicts protein aggregation by modeling intermolecular beta-sheet formation to identify aggregation-prone regions relevant to protein misfolding diseases.

Key Features:

• Predictive Accuracy: Validated on a dataset of 179 literature peptides and an additional 71 peptides derived from human disease-related proteins, including prion protein, lysozyme, and beta2-microglobulin.
• Disease Relevance: Predicts pathogenic and protective mutations in disease-associated proteins such as the Alzheimer’s beta-peptide, human lysozyme, and transthyretin.
• Discrimination Between Propensity and Aggregation: Distinguishes intrinsic beta-sheet propensity from actual aggregation propensity to provide more specific predictions of aggregation-prone sequences.
• Automated Design Strategy: Provides sequence-based predictions to guide modification of protein aggregation properties for engineering or experimental studies.

Scientific Applications:

• Protein Misfolding Diseases: Identification of aggregation-prone regions in proteins implicated in neurodegenerative disorders such as Alzheimer’s disease and prion diseases.
• Protein Engineering: Guiding sequence modifications to reduce or alter aggregation propensity for biotechnological and therapeutic protein design.
• Mutation Analysis: Assessing the effects of specific mutations on aggregation propensity to distinguish pathogenic versus protective variants.

Methodology:

TANGO employs a statistical mechanics framework that integrates physico-chemical principles of beta-sheet formation, models aggregation assuming core regions within aggregates are fully buried, and was evaluated on the described peptide datasets.