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

Developing optimal non-linear scoring function for protein design.

Changyu Hu1, Xiang Li, Jie Liang

  • 1Department of Bioengineering, SEO, MC-063, University of Illinois at Chicago, 851 S. Morgan Street, Room 218, Chicago, IL 60607-7052, USA.

Bioinformatics (Oxford, England)
|June 26, 2004
PubMed
Summary

Developing an advanced protein design scoring function is crucial for identifying compatible protein sequences. Our new non-linear Gaussian kernel function perfectly distinguishes native proteins from millions of decoys, outperforming linear methods.

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

Unraveling the Distinct Roles of Al and Ca in Microstructure Evolution and Tensile Response of Extruded Mg-Al-Ca Alloys.

Materials (Basel, Switzerland)·2026
Same author

Harnessing a Functional Rhizobial Partnership with <i>Astragalus sinicus</i> L. to Unlock Phytoremediation Potential for Soil Heavy Metals.

Journal of agricultural and food chemistry·2026
Same author

150 Gbaud high-efficiency O-band silicon microring modulator.

Optics express·2026
Same author

A social network analysis of fraud prediction on crowdsourcing platforms.

PloS one·2026
Same author

Nonvolatile PCM-driven photonic computing using programmable sub-wavelength digital metasurfaces.

Optics letters·2025
Same author

MOF-derived NiAl<sub>2</sub>O<sub>4</sub>/NiCo<sub>2</sub>O<sub>4</sub> porous materials as supercapacitors with high electrochemical performance.

Physical chemistry chemical physics : PCCP·2024

Area of Science:

  • Computational biology
  • Protein engineering
  • Bioinformatics

Background:

  • Protein design requires accurate scoring functions to identify sequences compatible with specific folds.
  • Current methods face challenges in characterizing the global fitness landscape of diverse proteins.

Purpose of the Study:

  • To develop and evaluate a novel, non-linear scoring function for protein sequence design.
  • To improve the accuracy of distinguishing native protein sequences from decoys.

Main Methods:

  • Utilized two geometric views and a mixture of non-linear Gaussian kernel functions.
  • Formulated a simplified protein sequence design problem to test the scoring function.
  • Evaluated performance against a large dataset of native proteins and sequence decoys.

Related Experiment Videos

Main Results:

  • The proposed scoring function achieved perfect discrimination between 440 native proteins and 14 million sequence decoys.
  • Demonstrated that linear scoring functions are inadequate for this task.
  • In a blind test, misclassified only 13 out of 194 unrelated proteins, significantly outperforming existing linear methods.

Conclusions:

  • Non-linear scoring functions, specifically Gaussian kernel mixtures, are superior for protein sequence design.
  • The developed method offers a significant advancement in accurately predicting protein sequence compatibility.
  • This approach has implications for developing more effective protein folding scoring functions.