QUARK

QUARK predicts three-dimensional protein structures from amino acid sequences using ab initio fragment assembly for protein structure modeling.


Key Features:

  • Ab initio folding: Performs protein structure prediction solely from amino acid sequences using ab initio approaches.
  • Fragment assembly: Assembles tertiary structures using structural motifs extracted from previously solved proteins.
  • Gapless-threading: Employs a gapless-threading technique to generate position-specific structure fragments.
  • Statistical consistency analysis: Derives position-specific fragments through statistical consistency analysis.
  • Distance profiles: Extracts distance profiles from derived fragments for use in structure modeling.
  • Torsion angle pairs: Extracts torsion angle pairs from derived fragments as modeling restraints.
  • Optimal fragment length: Uses an optimal fragment length of approximately 10 residues for structural assembly.
  • Fragment count requirement: Requires at least 100 fragments at each location to achieve optimal assembly.
  • Comparative accuracy: Achieves accuracy comparable to machine-learning-based methods, including when using fragments from unrelated proteins.

Scientific Applications:

  • Drug discovery: Provides predicted protein structures from sequence data to support research in drug discovery.
  • Understanding disease mechanisms: Supplies structural models to assist studies of molecular bases of disease.
  • Enzyme design: Generates three-dimensional models that can inform enzyme design and engineering efforts.
  • Structural biology: Enables ab initio modeling of protein tertiary structure from sequence data for structural biology research.

Methodology:

Applies gapless-threading to generate position-specific structure fragments; derives fragments via statistical consistency analysis to extract distance profiles and torsion angle pairs; performs fragment assembly using structural motifs from previously solved proteins; uses an optimal fragment length of ~10 residues and at least 100 fragments per location.

Topics

Details

Tool Type:
web application
Operating Systems:
Linux, Windows, Mac
Added:
8/3/2017
Last Updated:
11/25/2024

Operations

Publications

Xu D, Zhang Y. Toward optimal fragment generations for <i>ab initio</i> protein structure assembly. Proteins: Structure, Function, and Bioinformatics. 2012;81(2):229-239. doi:10.1002/prot.24179. PMID:22972754. PMCID:PMC3551984.

PMID: 22972754
PMCID: PMC3551984
Funding: - NSF Career Award: DBI 1027394 - National Institute of General Medical Sciences: GM083107, GM084222

Documentation

Links