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

Separating model optimization and model validation in statistical cross-validation as applied to crystallography.

Gerard J Kleywegt1

  • 1Department of Cell and Molecular Biology, Uppsala University, Biomedical Centre, Box 596, SE-751 24 Uppsala, Sweden. gerard@xray.bmc.uu.se

Acta Crystallographica. Section D, Biological Crystallography
|August 21, 2007
PubMed
Summary
This summary is machine-generated.

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

Chart builder: an interactive tool for user driven data visualization in the electron microscopy data bank.

Frontiers in bioinformatics·2026
Same author

EMICSS: added-value annotations for EMDB entries.

Bioinformatics advances·2025
Same author

MIFA: Metadata, Incentives, Formats and Accessibility guidelines to improve the reuse of AI datasets for bioimage analysis.

Nature methods·2025
Same author

Ten recommendations for organising bioimaging data for archival.

F1000Research·2024
Same author

Community recommendations on cryoEM data archiving and validation.

IUCrJ·2024
Same author

Community recommendations on cryoEM data archiving and validation: Outcomes of a wwPDB/EMDB workshop on cryoEM data management, deposition and validation.

ArXiv·2023

Introducing a dormant hold-out set in macromolecular crystallography avoids bias in model refinement. This small modification to cross-validation ensures R(free) remains an independent check of model quality.

Area of Science:

  • Crystallography
  • Structural Biology
  • Computational Biology

Background:

  • Statistical cross-validation is crucial for refining macromolecular crystallography models.
  • The free R value (R(free)) assesses model quality using a test set of reflections.
  • Current methods use the same test set for both model optimization and validation, potentially introducing bias.

Purpose of the Study:

  • To address potential bias in macromolecular model refinement.
  • To propose a method to maintain the independence of the R(free) metric.
  • To enhance the reliability of structural model validation in crystallography.

Main Methods:

  • Implementing a dormant hold-out set of reflections.
  • Modifying the standard cross-validation protocol slightly.

Related Experiment Videos

  • Ensuring the hold-out set is not used during model parameterization.

    • Main Results:

    • The proposed method effectively avoids bias associated with using the same test set for optimization and validation.
    • R(free) remains an independent and reliable indicator of model quality.
    • The modification requires minimal changes to existing cross-validation procedures.

    Conclusions:

    • A dormant hold-out set is a simple yet effective strategy to prevent bias in crystallographic model refinement.
    • This approach preserves the integrity of R(free) as a true measure of model accuracy.
    • The method offers a practical improvement for validating macromolecular structures.