EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

GKNnet: an relational graph convolutional network-based method with knowledge-augmented activation layer for microbial structural variation detection

Fengyi Guo, Yuanbo Li, Hongyuan Zhao, Xiaogang Liu, Jian Mao iD, Dongna Ma, Shuangping Liu

Briefings in Bioinformatics · 2025

DOI: 10.1093/bib/bbaf200 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2025. Verfasst von Fengyi Guo und 6 weiteren Autor:innen. Beteiligte Einrichtung: School of Artificial Intelligence and Computer Science, Jiangnan University, 1800 Lihu Avenue, Binhu District, Wuxi, Jiangsu 214122, China. Hauptthemen laut MEDLINE: Genomic Structural Variation, Computational Biology, Neural Networks, Computer, Genome, Microbial. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundamental Research Funds for the Central Universities, National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Structural variants (SVs) in microbial genomes play a critical role in phenotypic changes, environmental adaptation, and species evolution, with deletion variations particularly closely linked to phenotypic traits. Therefore, accurate and comprehensive identification of deletion variations is essential. Although long-read sequencing technology can detect more SVs, its high error rate introduces substantial noise, leading to high false-positive and low recall rates in existing SV detection algorithms. This paper presents an SV detection method based on graph convolutional networks (GCNs). The model first represents node features through a heterogeneous graph, leveraging the GCN to precisely identify variant regions. Additionally, a knowledge-augmented activation layer (KANLayer) with a learnable activation function is introduced to reduce noise around variant regions, thereby improving model precision and reducing false positives. A clustering algorithm then aggregates multiple overlapping regions near the variant center into a single accurate SV interval, further enhancing recall. Validation on both simulated and real datasets demonstrates that our method achieves superior F1 scores compared to benchmark methods (cuteSV, Sniffles, Svim, and Pbsv), highlighting its advantage and robustness in SV detection and offering an innovative solution for microbial genome structural variation research.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Genomic Structural VariationComputational BiologyNeural Networks, ComputerGenome, MicrobialAlgorithms

Stichwörter der Autor:innen

graph neural networkheterogeneous graphlong-read sequencingstructural variation detection

Autor:innen und Institutionen

  • School of Artificial Intelligence and Computer Science, Jiangnan University, 1800 Lihu Avenue, Binhu District, Wuxi, Jiangsu 214122, China.
  • National Engineering Research Center of Cereal Fermentation and Food Biomanufacturing, State Key Laboratory of Food Science and Technology, School of Food Science and Technology, Jiangnan University, 1800 Lihu Avenue, B…
  • Luzhou Laojiao Group Co. Ltd, 157 Guojiao Road, Jiangyang District, Luzhou 646000, Sichuan, China.
  • Shaoxing Key Laboratory of Traditional Fermentation Food and Human Health, Jiangnan University (Shaoxing) Industrial Technology Research Institute, Keqiao District, Shaoxing 312000, Zhejiang, China.

ORCID-Profile: Jian Mao

Förderung

  • Fundamental Research Funds for the Central Universities
  • National Natural Science Foundation of China

Zitieren

APA

Guo, F., Li, Y., Zhao, H., Liu, X., Mao, J., Ma, D., & Liu, S. (2025). GKNnet: an relational graph convolutional network-based method with knowledge-augmented activation layer for microbial structural variation detection. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbaf200

Vancouver

Guo F, Li Y, Zhao H, Liu X, Mao J, Ma D, et al. GKNnet: an relational graph convolutional network-based method with knowledge-augmented activation layer for microbial structural variation detection. Briefings in Bioinformatics. 2025. doi:10.1093/bib/bbaf200

BibTeX
@article{Guo2025,
  author = {Guo, Fengyi and Li, Yuanbo and Zhao, Hongyuan and Liu, Xiaogang and Mao, Jian and Ma, Dongna and Liu, Shuangping},
  title = {GKNnet: an relational graph convolutional network-based method with knowledge-augmented activation layer for microbial structural variation detection},
  journal = {Briefings in Bioinformatics},
  year = {2025},
  doi = {10.1093/bib/bbaf200}
}
Kostenlos starten →