EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GENNUS: generative approaches for nucleotide sequences enhance mirtron classification

Alisson Gaspar Chiquitto iD, Liliane Santana Oliveira, Pedro Henrique Bugatti, Priscila Tiemi Maeda Saito, Mark Basham, Roberto Tadeu Raittz, Alexandre Rossi Paschoal iD

NAR Genomics and Bioinformatics · 2025

DOI: 10.1093/nargab/lqaf072 ↗

Kurz gefasst

Fachartikel in der Zeitschrift NAR Genomics and Bioinformatics, 2025. Verfasst von Alisson Gaspar Chiquitto und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Federal University of Technology of Paraná-UTFPR, Cornélio Procópio, 86300-000, Brazil. Hauptthemen laut MEDLINE: MicroRNAs. Die Arbeit nennt 50 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Programa de Afastamento para Capacitação em Curso de Pós-Graduação de Docentes, Federal Institute of Education Science and Technology of Mato Grosso do Sul/Brasil, Conselho Nacional de Desenvolvimento Científico e Tecnológico.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Classifying non-coding RNA (ncRNA) sequences, particularly mirtrons, is essential for elucidating gene regulation mechanisms. However, the prevalent class imbalance in ncRNA datasets presents significant challenges, often resulting in overfitting and diminished generalization in machine learning models. In this study, GENNUS (GENerative approaches for NUcleotide Sequences) is proposed, introducing novel data augmentation strategies using generative adversarial networks (GANs) and synthetic minority over-sampling technique (SMOTE) to enhance mirtron and canonical microRNA (miRNA) classification performance. Our GAN-based methods effectively generate high-quality synthetic data that capture the intricate patterns and diversity of real mirtron sequences, eliminating the need for extensive feature engineering. Through four experiments, it is demonstrated that models trained on a combination of real and GAN-generated data improve classification accuracy compared to traditional SMOTE techniques or only with real data. Our findings reveal that GANs enhance model performance and provide a richer representation of minority classes, thus improving generalization capabilities across various machine learning frameworks. This work highlights the transformative potential of synthetic data generation in addressing data limitations in genomics, offering a pathway for more effective and scalable mirtron and canonical miRNA classification methodologies. GENNUS is available at https://github.com/chiquitto/GENNUS; and https://doi.org/10.6084/m9.figshare.28207328.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

MicroRNAsMachine LearningHumansSoftwareAlgorithms

Substanzen

MicroRNAs

Autor:innen und Institutionen

  • Department of Computer Science, Federal University of Technology of Paraná-UTFPR, Cornélio Procópio, 86300-000, Brazil.
  • Department of Informatics, Federal Institute of Education, Science and Technology of Mato Grosso do Sul - IFMS, Naviraí, 79947-334, Brazil.
  • Department of Computing, Federal University of São Carlos-UFSCar, São Carlos, 13565-905, Brazil.
  • The Rosalind Franklin Institute, Didcot, OX11 0QX, United Kingdom.
  • Laboratory of Bioinformatics, Professional and Technological Education Sector, Federal University of Paraná-UFPR, São Paulo, 81520-260, Brazil.

ORCID-Profile: Alisson Gaspar Chiquitto, Alexandre Rossi Paschoal

Förderung

  • Programa de Afastamento para Capacitação em Curso de Pós-Graduação de Docentes
  • Federal Institute of Education Science and Technology of Mato Grosso do Sul/Brasil
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico
  • Fundação Araucária
  • The Rosalind Franklin Institute
  • Coordenação de Aperfeiçoamento de Pessoal de Nível Superior

Zitieren

APA

Gaspar Chiquitto, A., Oliveira, L. S., Bugatti, P. H., Saito, P. T. M., Basham, M., Raittz, R. T., & Paschoal, A. R. (2025). GENNUS: generative approaches for nucleotide sequences enhance mirtron classification. NAR Genomics and Bioinformatics. https://doi.org/10.1093/nargab/lqaf072

Vancouver

Gaspar Chiquitto A, Oliveira LS, Bugatti PH, Saito PTM, Basham M, Raittz RT, et al. GENNUS: generative approaches for nucleotide sequences enhance mirtron classification. NAR Genomics and Bioinformatics. 2025. doi:10.1093/nargab/lqaf072

BibTeX
@article{GasparChiquitto2025,
  author = {Gaspar Chiquitto, Alisson and Oliveira, Liliane Santana and Bugatti, Pedro Henrique and Saito, Priscila Tiemi Maeda and Basham, Mark and Raittz, Roberto Tadeu and Paschoal, Alexandre Rossi},
  title = {GENNUS: generative approaches for nucleotide sequences enhance mirtron classification},
  journal = {NAR Genomics and Bioinformatics},
  year = {2025},
  doi = {10.1093/nargab/lqaf072}
}
Kostenlos starten →