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

Selecting high-quality negative samples for effectively predicting protein-RNA interactions.

Zhanzhan Cheng1, Kai Huang1, Yang Wang2

  • 1School of Computer Science, Fudan University, Handan Road, Shanghai, 200433, China.

BMC Systems Biology
|April 1, 2017
PubMed
Summary

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

Response of Tomicus yunnanensis (Coleoptera: Scolytinae) to infested and uninfested Pinus yunnanensis bolts.

Journal of economic entomology·2010
Same author

Construction of novel tumor necrosis factor-alpha mutants with reduced toxicity and higher cytotoxicity on human tumor cells.

Science in China. Series C, Life sciences·2010
Same author

Impact of MSW landfill on the environmental contamination of phthalate esters.

Waste management (New York, N.Y.)·2010
Same author

[Comparison of Cj1136, Cj1138 and Cj1139 genes among Campylobacter jejuni strains].

Zhonghua liu xing bing xue za zhi = Zhonghua liuxingbingxue zazhi·2010
Same author

Alveolar stability under different combinations of positive end-expiratory pressure and tidal volume: alveolar microscopy in isolated injured rat lungs.

Chinese medical journal·2010
Same author

[Research on permeability of landfill's body in primary compression].

Huan jing ke xue= Huanjing kexue·2010
Same journal

Correction to: A quantitative systems pharmacology (QSP) model for Pneumocystis treatment in mice.

BMC systems biology·2019
Same journal

Predicting disease-related phenotypes using an integrated phenotype similarity measurement based on HPO.

BMC systems biology·2019
Same journal

Fusing gene expressions and transitive protein-protein interactions for inference of gene regulatory networks.

BMC systems biology·2019
Same journal

A fast and efficient count-based matrix factorization method for detecting cell types from single-cell RNAseq data.

BMC systems biology·2019
Same journal

GNE: a deep learning framework for gene network inference by aggregating biological information.

BMC systems biology·2019
Same journal

FCMDAP: using miRNA family and cluster information to improve the prediction accuracy of disease related miRNAs.

BMC systems biology·2019
See all related articles
This summary is machine-generated.

Developing reliable negative samples significantly enhances machine learning models for predicting Protein-RNA Interactions (PRIs). This new method improves accuracy and performance compared to random sampling, aiding in understanding cellular activities.

Area of Science:

  • Bioinformatics
  • Computational Biology
  • Molecular Biology

Background:

  • Protein-RNA Interactions (PRIs) are crucial for cellular functions.
  • Current machine learning methods for PRI identification show unsatisfactory performance.
  • A key limitation is the use of unreliable negative samples during training.

Purpose of the Study:

  • To propose a novel method for generating reliable negative samples to improve PRI prediction.
  • To enhance the performance of machine learning models in identifying Protein-RNA Interactions.

Main Methods:

  • Collected known PRIs as positive samples.
  • Generated reliable negative samples using a novel method and compared with random methods.
  • Created 18 diverse datasets across species and sample ratios.
Keywords:
Protein-RNA interactionsReliable negative samplesUnreliable negative samples

Related Experiment Videos

  • Extracted sequence-based features and reduced dimensionality using a filter-based approach.
  • Evaluated Support Vector Machine (SVM), Random Forest (RF), and Naive Bayes (NB) classifiers.

    • Main Results:

    • Using the proposed method for negative sample generation led to substantial performance improvements across all tested classifiers compared to random sampling.
    • Accuracy and geometric mean improvements were significant, reaching up to 204.5% and 68.7% for SVM, 174.5% and 53.9% for RF, and 80.9% and 54.3% for NB.

    Conclusions:

    • The developed method for generating reliable negative samples is effective in boosting the performance of PRI identification.
    • This approach offers a valuable tool for improving the accuracy of computational methods in predicting Protein-RNA Interactions.