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

A-Prot: protein structure modeling using MSA transformer.

Yiyu Hong1, Juyong Lee2,3, Junsu Ko1

  • 1Arontier Co, Seoul, Republic of Korea.

BMC Bioinformatics
|March 17, 2022
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

The U-CLEAR study: Uncovering hemodynamic and vascular contributors to chronic refractory lower extremity edema.

Phlebology·2026
Same author

Discovery of a KLHL41 Ligand for Muscle Specific Protein Degradation.

Nature communications·2026
Same author

A snare technique for valve-in-valve transcatheter aortic valve replacement in a stented coarctation of the aorta: a case report.

Frontiers in cardiovascular medicine·2026
Same author

BA-Pred and RMSD-Pred: Integrated Graph Neural Network Models for Accurate Protein-Ligand Binding Affinity and Binding Pose Prediction.

Journal of chemical information and modeling·2026
Same author

Docking of millions: accelerating a million-scale virtual screening using deep learning.

Briefings in bioinformatics·2026
Same author

Hydralazine-Associated Pericardial Effusion Causing Cardiac Tamponade Without Hypotension.

JACC. Case reports·2026
Same journal

OpenIMC: an open-source platform for analyzing single-cell and spatial proteomics by imaging mass cytometry.

BMC bioinformatics·2026
Same journal

NAP: an open source pipeline for cross-domain microbiome profiling using Nanopore sequencing-derived amplicon data.

BMC bioinformatics·2026
Same journal

SurvGME: an R package for survival analysis with graphical and measurement error models.

BMC bioinformatics·2026
Same journal

SimMapNet: a Bayesian framework for gene regulatory network inference using gene ontology similarities as external hint.

BMC bioinformatics·2026
Same journal

Dual channel drug-drug interactions extraction based on cross attention.

BMC bioinformatics·2026
Same journal

FeSseqdb: a curated sequence-level database and interpretable machine learning framework for identifying iron-sulfur proteins.

BMC bioinformatics·2026
See all related articles

A new protein structure prediction method, A-Prot, uses evolutionary information from multiple sequence alignments to generate accurate 3D models. This lighter model offers a cost-effective alternative to existing deep learning approaches.

Area of Science:

  • Computational Biology
  • Structural Biology
  • Deep Learning

Background:

  • Deep learning significantly advanced protein 3D structure prediction accuracy.
  • AlphaFold (AF) achieved near-experimental accuracy, highlighting the value of evolutionary information in multiple sequence alignments (MSAs).
  • AF's training demands high computational resources, necessitating development of lighter prediction models.

Purpose of the Study:

  • To develop a novel, computationally efficient protein 3D structure modeling method.
  • To leverage state-of-the-art protein language models for improved prediction accuracy.

Main Methods:

  • Proposed A-Prot, a protein 3D structure modeling method utilizing MSA Transformer.
  • Extracted MSA feature tensor and row attention maps.
Keywords:
Deep learningMultiple sequence alignmentProtein language modelProtein structure prediction

Related Experiment Videos

  • Converted features into 2D residue-residue distance and dihedral angle predictions.

    • Main Results:

    • A-Prot demonstrated superior prediction of long-range contacts compared to existing methods.
    • Modeled 3D structures for CASP14 free modeling and hard template-based targets.
    • A-Prot models exhibited higher accuracy than most top CASP14 server groups.

    Conclusions:

    • A-Prot effectively captures protein evolutionary and structural information at a lower computational cost.
    • The method shows promise for developing other protein property prediction tools.