Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Conformational variability of solution nuclear magnetic resonance structures

A M Bonvin1, A T Brünger

  • 1Howard Hughes Medical Institute, Yale University, New Haven, CT 06520, USA.

Journal of Molecular Biology
|June 30, 1995
PubMed
Summary

Cross-validation in nuclear magnetic resonance (NMR) data refinement prevents overfitting. This method accurately determined whether single or multiple conformer models best represent protein structures, like interleukin 4 and 8.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Time- and ensemble-averaged direct NOE restraints.

Journal of biomolecular NMR·2012
Same author

A smooth and differentiable bulk-solvent model for macromolecular diffraction.

Acta crystallographica. Section D, Biological crystallography·2010
Same author

Climate change and the integrity of science.

Science (New York, N.Y.)·2010
Same author

Highly specific interactions between botulinum neurotoxins and synaptic vesicle proteins.

Cellular and molecular life sciences : CMLS·2008
Same author

Structural characterization of the intramolecular interaction between the SH3 and guanylate kinase domains of PSD-95.

Molecular cell·2002
Same author

Rapid protein fold determination using secondary chemical shifts and cross-hydrogen bond 15N-13C' scalar couplings (3hbJNC').

Journal of biomolecular NMR·2002

Area of Science:

  • Structural biology
  • Biophysics
  • Computational chemistry

Background:

  • Experimental data in X-ray crystallography and NMR spectroscopy are time and ensemble-averaged.
  • Refining ensembles against NMR data risks overfitting due to increased parameters.
  • Cross-validation offers an unbiased measure to assess model fit and prevent overfitting.

Purpose of the Study:

  • To apply complete cross-validation to nuclear Overhauser effect (NOE)-derived distance refinement.
  • To correlate cross-validation measures with the number of models representing conformational variability.
  • To develop a probability map-based method for generating accurate structural ensembles.

Main Methods:

  • Complete cross-validation applied to NOE-derived distance refinement.

Related Experiment Videos

  • Generation of structural ensembles using a probability map method.
  • Application to synthetic data and experimental NMR data for interleukin 4 and 8.

    • Main Results:

    • For interleukin 4, cross-validation favored a single-conformer model, revealing overfitting with multiple conformers.
    • For interleukin 8, cross-validation identified a twin-conformer model as the most accurate representation.
    • Distinct conformations in the interleukin 8 loop (residues 16-22) were observed, supported by 15N relaxation studies.

    Conclusions:

    • Complete cross-validation is crucial for accurate structure determination from NMR data.
    • It effectively distinguishes between single and multiple conformer models, preventing overfitting.
    • The study provides insights into the conformational dynamics of interleukin 4 and 8, relevant to their functions.