BCL

BCL assembles de novo protein structures from idealized α-helices and β-strands to predict complex protein topologies and generate models from sparse or low-resolution experimental data.

Key Features:

• Topology Assembly: Constructs protein topologies by assembling idealized α-helices and β-strands to explore complex structural formations.
• Monte Carlo Metropolis Simulated Annealing: Applies Monte Carlo Metropolis simulated annealing to optimize a knowledge-based potential.
• Knowledge-based Potential Factors: The potential is informed by radius of gyration, β-strand pairing, secondary structure element (SSE) packing, amino acid pair distances, and environmental context.
• Optimization Parameters: Considers contact order, agreement with secondary structure predictions, and loop closure to improve model accuracy.
• Non-local Contact Sampling: Samples non-local contacts by discontinuing protein chains during simulations to facilitate complex topology formation.
• Flexible Loop Exclusion: Excludes flexible loop regions from folding simulations to reduce the conformational search space and accelerate sampling.

Scientific Applications:

• Modeling from sparse or low-resolution data: Generates structural models when experimental data are sparse or low resolution to support structural interpretation.
• Restraint generation for secondary structure regions: Produces models that provide restraints for defined secondary structure regions to assist downstream modeling or experimental refinement.

Methodology:

Assembles protein topologies from idealized α-helices and β-strands and uses Monte Carlo Metropolis simulated annealing to optimize a knowledge-based potential informed by radius of gyration, β-strand pairing, SSE packing, amino acid pair distances, and environmental context; the method considers contact order, agreement with secondary structure predictions, and loop closure, samples non-local contacts by discontinuing protein chains, and excludes flexible loop regions to accelerate simulations; benchmarked on 66 proteins (83–293 amino acids) achieving RMSD100 < 8.0 Å for 61 proteins, recovering >20% native contacts, and assembling topologies up to 215 residues with relative contact order 0.46.