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

Will my protein crystallize? A sequence-based predictor.

Pawel Smialowski1, Thorsten Schmidt, Jürgen Cox

  • 1Department of Genome Oriented Bioinformatics, Technische Universität München, Wissenschaftszentrum Weihenstephan, Freising, Germany.

Proteins
|November 30, 2005
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

A multi-center cross-sectional study to explore cancer patients' views on facilitators and barriers for the implementation of oncological exercise therapy.

Frontiers in oncology·2026
Same author

Proteomic analysis of dental enamel from 20 Homo naledi individuals shows no male markers.

Cell·2026
Same author

From mechanism to substratome: unraveling mysteries of γ-secretase.

The Journal of biological chemistry·2026
Same author

Effects of Transcutaneous Electrical Nerve Stimulation, Electrical Muscle Stimulation, Exercise and Heat on Back Pain-A Randomized Controlled Clinical Exploratory Trial.

European journal of pain (London, England)·2026
Same author

Improved Peptide Search for Identification of SUMO and Sequence-Based Modifiers, in MaxSBM.

Molecular & cellular proteomics : MCP·2026
Same author

Bimodal Peptide Collision Cross Section Distribution Reflects Two Stable Conformations in the Gas Phase.

Journal of proteome research·2026
Same journal

Engineered HSP90-MP65 Bivalent Fusion Antigen: A Novel Vaccine Candidate Against Invasive Candidiasis.

Proteins·2026
Same journal

Physics-Based Energy Functions for Computational Protein Design.

Proteins·2026
Same journal

Impact of Stabilizing Osmolytes on the Conformational Dynamics of Human and Rat Islet Amyloid Polypeptides.

Proteins·2026
Same journal

Stabilization of Bone Morphogenetic Protein-2 at Physiological pH: Contrasting Roles of CHAPS and Arginine in Aggregation Inhibition.

Proteins·2026
Same journal

Structural Insights Into the Function of Leishmania major Adenylosuccinate Lyase.

Proteins·2026
Same journal

Generalizing the Gaussian Network Model: Spanning-Tree Thermodynamics Shows Entropy-Driven KRAS Activation.

Proteins·2026
See all related articles

We developed a machine learning model to predict protein crystallizability using amino acid sequences. This tool aids in selecting proteins for structural genomics, improving efficiency in structural biology research.

Area of Science:

  • Structural Biology
  • Computational Biology
  • Biophysics

Background:

  • Protein structure determination is crucial for understanding biological function.
  • Predicting protein crystallizability aids in efficient structure-based drug design and functional studies.
  • Existing methods for predicting protein crystallizability have limitations.

Purpose of the Study:

  • To develop a machine learning approach for predicting protein crystallizability directly from amino acid sequences.
  • To leverage differences between proteins solved by X-ray crystallography and Nuclear Magnetic Resonance (NMR) spectroscopy.
  • To create a valuable tool for structural genomics to improve target selection.

Main Methods:

  • Utilized a machine learning approach based on sequence-based features.

Related Experiment Videos

  • Employed frequencies of mono-, di-, and tripeptides from standard and reduced amino acid alphabets.
  • Implemented a two-layered classification system: Support Vector Machines (SVM) followed by Naive Bayes.
  • Adjusted sequence length distributions to mitigate bias from protein size differences between datasets.
  • Incorporated cost-sensitive learning to handle unbalanced datasets.

    • Main Results:

    • Achieved an overall prediction accuracy of 67% in a 10-fold cross-validation.
    • Demonstrated specific accuracies of 65% for crystallizable proteins and 69% for non-crystallizable proteins.
    • The developed algorithm effectively handles unbalanced class representation in real-world datasets.

    Conclusions:

    • The proposed machine learning method provides a valuable tool for predicting protein crystallizability.
    • This approach can enhance the efficiency of target selection in structural genomics projects.
    • The SECRET web server is available for public use, facilitating protein crystallizability predictions.