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

Predicting intrinsic disorder from amino acid sequence.

Zoran Obradovic1, Kang Peng, Slobodan Vucetic

  • 1Center for Information Science and Technology, Temple University, Philadelphia, Pennsylvania 19122, USA. zoran@ist.temple.edu

Proteins
|October 28, 2003
PubMed
Summary

Predicting protein intrinsic disorder from amino acid sequences is feasible. Machine learning models show varying accuracy, with better performance on longer disordered segments than shorter ones.

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

Highly Compressible Si<sub>3</sub>N<sub>4</sub> Nanowire Aerogels with Thermal Stability up to 1400 °C.

ACS applied materials & interfaces·2026
Same author

Field assessment of radon exposure in three zirconium and titanium mineral processing plants: radon concentrations, individual progeny concentrations, size distributions, etc.

Radiation protection dosimetry·2026
Same author

Causal relationship between plasma metabolites and chronic regional pain: a Mendelian randomization study.

Metabolism open·2026
Same author

Preparation of radon-free thoron sources based on <sup>228</sup>Th separated from industrial thorium nitrate.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine·2026
Same author

Targeting FAM83D triggers tumor cell senescence via cGAS-STING signaling activation and reprograms TAMs to combat glioma.

Journal of experimental & clinical cancer research : CR·2026
Same author

CXCL7/CXCR2 in the paraventricular thalamus mediates obesity-related pain.

Journal of neuroinflammation·2026

Area of Science:

  • Biochemistry
  • Computational Biology
  • Structural Biology

Background:

  • Proteins can exist in intrinsically disordered or ordered states.
  • Accurate prediction of these states is crucial for understanding protein function.

Purpose of the Study:

  • To evaluate the accuracy of computational predictors for intrinsic protein disorder.
  • To investigate factors influencing prediction accuracy, particularly for short versus long disordered segments.

Main Methods:

  • Blind prediction of order and disorder for 42 proteins using six different computational predictors.
  • Analysis of predictor performance based on residue type (ordered, disordered) and segment length.

Main Results:

  • Predictor accuracies for ordered regions ranged from 77% to 91%.

Related Experiment Videos

  • Disordered segment prediction accuracy varied, with better performance on longer segments (>30 residues) compared to shorter ones.
  • Ensemble neural network predictors showed improved accuracy, especially on ordered and long disordered regions.

    • Conclusions:

    • Intrinsic protein disorder is predictable from amino acid sequences.
    • Current predictors exhibit disparities in accuracy between short and long disordered segments, likely due to training data and prediction methodologies.