LOBO

LOBO predicts loop conformations in protein structures using an ab initio divide-and-conquer algorithm to generate ranked alternative conformations for molecular simulation and structural analysis.

Key Features:

• Ab initio divide-and-conquer algorithm: Employs an ab initio divide-and-conquer strategy to model loop regions within protein structures.
• Recursive decomposition: Recursively decomposes target loops into smaller segments until they can be analytically compiled.
• Precalculated look-up tables: Uses precalculated look-up tables that encompass potential conformations for loop segments of varying lengths.
• Unrestricted loop length handling: The look-up tables enable handling of loops without restriction on their length.
• Ranked conformational outputs: Generates a ranked set of possible loop conformations as output.
• Computational performance: Produces predictions within 20–180 seconds on a standard desktop PC.
• Quality metrics: Assesses prediction quality using global root-mean-square deviation (RMSD), with reported top-prediction RMSDs of ~1.06 Å for three-residue loops and ~3.72 Å for eight-residue loops.

Scientific Applications:

• Starting conformations for simulations: Generates alternative starting conformations for complex molecular simulations.
• Protein dynamics and interactions: Aids exploration of protein dynamics and protein–protein or protein–ligand interactions via alternative loop conformations.
• Structure analysis and modeling: Supports protein structure analysis and loop modeling in computational studies.

Methodology:

Uses an ab initio divide-and-conquer approach that recursively decomposes loops and analytically compiles segment conformations from precalculated look-up tables.