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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Automated Chemical Profiling of Wine by Solution NMR Spectroscopy: A Demonstration for Outreach and Education.

Journal of chemical education·2026
Same author

Detergents alter the stability and lipid binding properties of the CD1d immunoreceptor.

Protein science : a publication of the Protein Society·2025
Same author

Peptide Prenylation Follows Divergent Substrate Engagement Rules.

Journal of the American Chemical Society·2025
Same author

Large protein-like leader peptides engage differently with RiPP halogenases and lanthionine synthetases.

Nature communications·2025
Same author

Structure-Guided Temporin L Analogs Development to Inhibit the Main Protease of SARS-CoV‑2.

ACS medicinal chemistry letters·2025
Same author

AlphaFold modeling uncovers global structural features of class I and class II fungal hydrophobins.

Protein science : a publication of the Protein Society·2025

Related Experiment Video

Updated: Jun 12, 2026

A Bilingual Computational Workflow for Identifying Potential PLK1 Inhibitors in American Sign Language and English
14:34

A Bilingual Computational Workflow for Identifying Potential PLK1 Inhibitors in American Sign Language and English

Published on: April 3, 2026

The LiPP Benchmark Set for Modeling Lipid-Protein Complexes: Comparison of Co-Folding and Docking Methods.

Li-Yen Yang1, Shreyas Gupta1,2, Lauren N Mullininx1,2

  • 1School of Chemistry and Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332, United States.

Journal of Chemical Information and Modeling
|June 10, 2026
PubMed
Summary

Computational tools struggle to predict lipid-protein interactions due to limited data. The new Lipid-Protein Poses (LiPP) benchmark reveals current AI and physics-based methods have shortcomings in modeling these crucial biological complexes.

More Related Videos

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules
10:58

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules

Published on: July 25, 2013

Incorporating Target Protein Structure Flexibility and Dynamics in Computational Drug Discovery Using Ensemble-Based Docking Analysis
08:49

Incorporating Target Protein Structure Flexibility and Dynamics in Computational Drug Discovery Using Ensemble-Based Docking Analysis

Published on: June 20, 2025

Related Experiment Videos

Last Updated: Jun 12, 2026

A Bilingual Computational Workflow for Identifying Potential PLK1 Inhibitors in American Sign Language and English
14:34

A Bilingual Computational Workflow for Identifying Potential PLK1 Inhibitors in American Sign Language and English

Published on: April 3, 2026

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules
10:58

Protein WISDOM: A Workbench for In silico De novo Design of BioMolecules

Published on: July 25, 2013

Incorporating Target Protein Structure Flexibility and Dynamics in Computational Drug Discovery Using Ensemble-Based Docking Analysis
08:49

Incorporating Target Protein Structure Flexibility and Dynamics in Computational Drug Discovery Using Ensemble-Based Docking Analysis

Published on: June 20, 2025

Area of Science:

  • Biochemistry and Structural Biology
  • Computational Biology and Bioinformatics

Background:

  • Lipid-protein interactions are fundamental to cellular processes, impacting medicine and biotechnology.
  • Predicting these interactions computationally lags behind other molecular complexes due to a scarcity of structural data.
  • Existing methods require rigorous evaluation on standardized datasets.

Purpose of the Study:

  • To introduce the Lipid-Protein Poses (LiPP) benchmark, a curated dataset of 311 lipid-protein complex structures.
  • To systematically evaluate the performance of state-of-the-art AI and physics-based computational tools for modeling lipid-protein complexes.
  • To identify limitations in current methods for predicting lipid-protein binding poses.

Main Methods:

  • Development of the LiPP benchmark dataset comprising diverse lipid-protein complex structures.
  • Systematic performance evaluation of AI tools (AlphaFold 3, Chai-1, RoseTTAFold AA, DiffDock-L) and a physics-based tool (AutoDock Vina).
  • Analysis of the physical plausibility and accuracy of predicted lipid binding poses using criteria like RMSD.

Main Results:

  • AI and physics-based tools exhibit varied performance in predicting lipid-protein binding poses.
  • Physics-based docking and some AI predictors better maintained structural constraints compared to others.
  • AlphaFold 3 showed the highest success rate (76.1% at RMSD < 2 Å), but performance decreased significantly for unseen structures (47.2%).
  • Significant differences were observed in the ability of tools to provide reliable confidence metrics.

Conclusions:

  • Current computational methods, both AI- and physics-based, have substantial limitations in accurately modeling lipid-protein interactions.
  • The LiPP benchmark highlights the need for improved algorithms that consider lipid characteristics like size, class, and flexibility.
  • Further development is required to advance computational tools for lipid biology research and drug discovery.