CPHmodels-3.0

CPHmodels-3.0 predicts protein three-dimensional (3D) structures from amino acid sequences using single-template homology modeling to generate structurally informative models for structural biology applications.

Key Features:

• Single-template homology modeling: Builds 3D models based on a single PDB template identified by sequence-profile comparisons.
• Hybrid scoring functions: Incorporates the fast profile-profile scoring function from CPHmodels-2.0 optimized for close homology modeling.
• Remote homology-modeling fallback: Applies a computationally intensive remote homology-modeling algorithm when no suitable PDB template is identified in the initial search.
• Benchmark performance (CASP8): In CASP8 the method produced models for 94% of targets (117/128), with 74% (87/117) classified as high-reliability models averaging 4.6 Å RMSD and low-reliability models averaging 9.3 Å RMSD.
• Computational efficiency: Typical server response time for most queries is under 20 minutes.

Scientific Applications:

• Drug Discovery: Provides predicted 3D structures of target proteins to support structure-based drug design and target assessment.
• Functional Annotation: Supplies structural models that can aid inference of protein function for uncharacterized sequences.
• Protein Engineering: Delivers structural templates to guide design or modification of proteins with desired properties.

Methodology:

Performs single-template homology modeling using a profile-profile scoring function from CPHmodels-2.0 and invokes a remote homology-modeling algorithm when no suitable PDB template is found.