EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

WheatGP, a genomic prediction method based on CNN and LSTM

Chunying Wang, Di Zhang, Yuexin Ma, Yonghao Zhao, Ping Liu iD, Xiang Li iD

Briefings in Bioinformatics · 04.03.2025 · 26 (2)

DOI: 10.1093/bib/bbaf191 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, online erschienen am 25.04.2025. Verfasst von Chunying Wang und 5 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Wheat Improvement, Shandong Agricultural University, 61 Daizong Street, Tai'an 271018, China. Hauptthemen laut MEDLINE: Triticum, Neural Networks, Computer, Genomics, Genome, Plant. Die Arbeit wurde bisher 21-mal zitiert (Crossref) und nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Shandong Province Postdoctoral Innovation Project, Shandong Provincial Key Research and Development Program Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Wheat plays a crucial role in ensuring food security. However, its complex genetic structure and trait variation pose significant challenges for breeding superior varieties. In this study, a genomic prediction method for wheat (WheatGP) is proposed. WheatGP is designed to improve the phenotype prediction accuracy by modeling both additive genetic effects and epistatic genetic effects. It is primarily composed of a convolutional neural network (CNN) module and a long short-term memory (LSTM) module. The multilayer CNNs within the CNN module focus on capturing short-range dependencies within the genomic sequence. Meanwhile, the LSTM module, with its unique gating mechanism, is designed to retain long-distance dependency relationships between gene loci in the features. Therefore, WheatGP could comprehensively extract multilevel features from genomic inputs. Compared to ridge regression best linear unbiased prediction (rrBLUP), extreme gradient boosting (XGBoost), support vector regression (SVR), and deep neural network genomic prediction (DNNGP), WheatGP demonstrates a clear advantage in terms of prediction accuracy. The prediction accuracy for wheat yield reaches 0.73, while the prediction accuracies for various agronomic traits range between 0.62 and 0.78. It also exhibits robust performance across other crop types and multi-omics datasets. In addition, SHapley Additive exPlanations (SHAP) is employed to evaluate the contributions of inputs to the predictive model. As a high-performance tool for genomic prediction in wheat, WheatGP opens up new possibilities for achieving efficient and optimized wheat breeding.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

TriticumNeural Networks, ComputerGenomicsGenome, PlantPlant BreedingPhenotypeModels, Genetic

Stichwörter der Autor:innen

genomic predictionlong short-term memoryphenotypewheat

Autor:innen und Institutionen

  • State Key Laboratory of Wheat Improvement, Shandong Agricultural University, 61 Daizong Street, Tai'an 271018, China.
  • Shandong Engineering Research Center of Agricultural Equipment Intelligentization, College of Mechanical and Electronic Engineering, Shandong Agricultural University, 61 Daizong Street, Tai'an 271018, China.
  • College of Life Sciences, Shandong Agricultural University, 61 Daizong Street, Tai'an 271018, China.

ORCID-Profile: Ping Liu, Xiang Li

Förderung

  • Shandong Province Postdoctoral Innovation Project
  • Shandong Provincial Key Research and Development Program Project

Zitieren

APA

Wang, C., Zhang, D., Ma, Y., Zhao, Y., Liu, P., & Li, X. (2025). WheatGP, a genomic prediction method based on CNN and LSTM. Briefings in Bioinformatics, 26(2). https://doi.org/10.1093/bib/bbaf191

Vancouver

Wang C, Zhang D, Ma Y, Zhao Y, Liu P, Li X. WheatGP, a genomic prediction method based on CNN and LSTM. Briefings in Bioinformatics. 2025;26(2). doi:10.1093/bib/bbaf191

BibTeX
@article{Wang2025,
  author = {Wang, Chunying and Zhang, Di and Ma, Yuexin and Zhao, Yonghao and Liu, Ping and Li, Xiang},
  title = {WheatGP, a genomic prediction method based on CNN and LSTM},
  journal = {Briefings in Bioinformatics},
  year = {2025},
  volume = {26},
  number = {2},
  doi = {10.1093/bib/bbaf191}
}
Kostenlos starten →