DISCO

DISCO computes three-dimensional structures of protein homo-oligomers with cyclic symmetry by combining orientational constraints from residual dipolar couplings (RDCs) and intermolecular distance restraints such as NOE interactions or disulfide bonds to satisfy both orientational and distance constraints.

Key Features:

• Integration of RDC and distance restraints: DISCO combines orientational constraints from RDCs with intermolecular distance restraints (NOE interactions, disulfide bonds) to determine oligomer geometry.
• Graphical annulus representation: Intermolecular distance restraints are represented as annuli in a plane that encodes parameters of the symmetry axis, enabling geometric analysis via annulus overlap.
• Handling subunit ambiguity: The framework supports ambiguously assigned distance restraints arising from degenerate chemical shifts in symmetric homo-oligomers.
• Algorithmic efficiency: The problem is reduced to computing geometric arrangements of unions of annuli with expected time complexity O(n²), where n is the number of distance restraints.
• Guaranteed exhaustive solutions: The method identifies all oligomer structures consistent with both distance and orientational restraints rather than relying on stochastic sampling.

Scientific Applications:

• Trimeric E. coli diacylglycerol kinase: DISCO has been applied to determine the structure of the trimeric E. coli diacylglycerol kinase using RDCs and intermolecular distance restraints.
• Dimeric mutant of streptococcal protein G domain B1: DISCO has been applied to a dimeric mutant of the immunoglobulin-binding domain B1 of streptococcal protein G to compute oligomeric structure from ambiguous restraints.

Methodology:

The algorithm enforces that the oligomer symmetry axis is parallel to an eigenvector of the RDC alignment tensor, represents intermolecular distance restraints as annuli in a plane encoding symmetry-axis parameters, and computes geometric arrangements and overlaps of unions of annuli to enumerate all structures satisfying the experimental constraints with expected O(n²) time complexity.