FOLD-RATE

FOLD-RATE predicts protein folding rates from amino acid sequences to estimate folding kinetics using sequence-derived properties and structural class information.

Key Features:

• Amino Acid Properties: Utilizes physical-chemical, energetic, and conformational amino acid properties to correlate sequence features with folding rates.
• Structural Class Correlation: Classifies proteins into all-alpha, all-beta, and mixed classes and correlates these classes with folding rates of two- and three-state proteins, emphasizing the role of native-state topology.
• Linear Regression Models: Employs simple linear regression models incorporating amino acid properties and structural class information, reporting correlation coefficients of 0.99 (all-alpha), 0.96 (all-beta), and 0.95 (mixed-class) between predicted and experimental folding rates.
• Long-Range Order Parameter: Introduces a long-range order parameter that quantifies long-range interactions to improve folding-rate prediction accuracy across globular protein classes.

Scientific Applications:

• Protein Folding Studies: Predicts folding rates from sequence to aid investigation of folding kinetics and mechanisms relevant to protein misfolding diseases.
• Structural Biology Research: Supports analysis of how amino acid composition and sequence influence protein native-state topology, structure, and stability.
• Biotechnological Applications: Informs design and engineering of proteins by providing predicted folding-rate information relevant for industrial and applied contexts.

Methodology:

Uses simple linear regression and multiple regression analyses combining sequence-derived properties (e.g., beta-strand tendency, enthalpy change) with structural parameters (contact order, total contact distance); models are validated against experimental folding-rate data.