NetSurfP-3.0

NetSurfP-3.0 predicts residue-level protein structural features, including solvent accessibility, secondary structure, structural disorder, and backbone dihedral angles, to support structural analysis and protein engineering.

Key Features:

• Predictive Capabilities: Predicts solvent accessibility, secondary structure, structural disorder, and backbone dihedral angles for each residue in an amino acid sequence.
• Performance Improvements: Uses pre-trained protein language models to reduce runtime by up to 600 times compared to other commonly available methods while maintaining state-of-the-art prediction accuracy.

Scientific Applications:

• Structural Biology: Enhances understanding of protein folding and stability through residue-level structural annotations.
• Biotechnology: Assists in the design and engineering of proteins by providing per-residue structural information.
• Drug Discovery: Aids identification of potential drug targets by supplying structural insights at the residue level.

Methodology:

Integrates pre-trained protein language models within a machine-learning framework to process large sequence datasets efficiently, reducing computational runtime and achieving high accuracy on independent test datasets.