US-align: universal structure alignments of proteins, nucleic acids, and macromolecular complexes

Area of Science:

• Structural Biology
• Bioinformatics
• Computational Biology

Background:

• Structure comparison and alignment are critical in structural biology.
• Existing methods often struggle with aligning diverse macromolecular structures (proteins, RNAs, DNAs) uniformly.
• A need exists for a versatile tool capable of handling both monomeric and complex structures.

Purpose of the Study:

• To develop the first universal platform, US-align, for uniform structure alignment.
• To enable the alignment of monomer and complex structures across different macromolecule types (proteins, RNAs, DNAs).
• To facilitate heterogeneous oligomer structure comparison and template-based docking.

Main Methods:

• Development of a universal platform (US-align) based on a unified TM-score objective function.
• Integration of a heuristic alignment searching algorithm for efficient optimization.
• Large-scale benchmarks to evaluate performance against state-of-the-art methods.

Main Results:

• US-align demonstrated consistent advantages over existing methods in pairwise and multiple structure alignments.
• The platform achieved significant improvements in both accuracy and speed of structural alignment.
• The unified protocol effectively handles diverse molecular and structural types.

Conclusions:

• US-align provides a powerful and versatile solution for macromolecular structure comparison and alignment.
• The optimization of the unified objective function and efficient heuristic search are key to its performance.
• The platform facilitates advanced applications like heterogeneous oligomer analysis and template-based docking.