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

Protein secondary structure prediction for a single-sequence using hidden semi-Markov models.

Zafer Aydin1, Yucel Altunbasak, Mark Borodovsky

  • 1School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA 30332-0250, USA. aydinz@ece.gatech.edu

BMC Bioinformatics
|March 31, 2006
PubMed
Summary

This study enhances single-sequence protein secondary structure prediction using improved hidden semi-Markov models (HSMM). The new method, IPSSP, achieves comparable or better accuracy than existing algorithms for proteins lacking homologous sequences.

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

GraphUnet-SS: A novel deep learning model for protein secondary structure prediction based on U-Net architecture.

Computers in biology and medicine·2026
Same author

Translon: a single term for translated regions.

Nature methods·2025
Same author

GeneMark-ETP significantly improves the accuracy of automatic annotation of large eukaryotic genomes.

Genome research·2024
Same author

BRAKER3: Fully automated genome annotation using RNA-seq and protein evidence with GeneMark-ETP, AUGUSTUS, and TSEBRA.

Genome research·2024
Same author

An effective colorectal polyp classification for histopathological images based on supervised contrastive learning.

Computers in biology and medicine·2024
Same author

BRAKER3: Fully automated genome annotation using RNA-seq and protein evidence with GeneMark-ETP, AUGUSTUS and TSEBRA.

bioRxiv : the preprint server for biology·2023

Area of Science:

  • Computational Biology
  • Bioinformatics
  • Structural Biology

Background:

  • Protein secondary structure prediction accuracy is approaching its theoretical limit.
  • Single-sequence prediction methods are crucial for proteins without known homologs.
  • Current single-sequence methods lack the accuracy of those utilizing homologous protein information.

Purpose of the Study:

  • To refine and extend the hidden semi-Markov model (HSMM) for improved single-sequence protein secondary structure prediction.
  • To develop novel residue dependency models and training methods for enhanced prediction accuracy.

Main Methods:

  • Refinement and extension of the hidden semi-Markov model (HSMM).
  • Introduction of improved residue dependency models considering amino acid correlations at structural segment borders.

Related Experiment Videos

  • Implementation of an iterative training method for HSMM parameter estimation.

    • Main Results:

    • The new method, IPSSP, demonstrates accuracy comparable or superior to BSPSS and PSIPRED under single-sequence conditions.
    • Improved residue dependency models and specialized HSMMs contribute to enhanced prediction performance.
    • Iterative training refines HSMM parameters, leading to better accuracy measures.

    Conclusions:

    • New dependency models and training methods significantly improve single-sequence protein secondary structure prediction.
    • Results validated under cross-validation using a dataset with no highly similar sequence pairs.
    • Future advancements hold promise for improved function prediction in orphan proteins.