EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

CRISPR-MFH: A Lightweight Hybrid Deep Learning Framework with Multi-Feature Encoding for Improved CRISPR-Cas9 Off-Target Prediction

Yanyi Zheng iD, Quan Zou iD, Jian Li, Yanpeng Yang iD

Genes · 28.03.2025 · 16 (4) · 387

DOI: 10.3390/genes16040387 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Genes, online erschienen am 28.03.2025. Verfasst von Yanyi Zheng und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Landscape Architecture, Beijing Forestry University, Beijing 100083, China. Hauptthemen laut MEDLINE: Deep Learning, CRISPR-Cas Systems, Gene Editing. Die Arbeit wurde bisher 12-mal zitiert (Crossref) und nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, scientific research project of the education department of Zhejiang province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

The CRISPR-Cas9 system has emerged as one of the most promising gene-editing technologies in biology. However, off-target effects remain a significant challenge. While recent advances in deep learning have led to the development of models for off-target prediction, these models often fail to fully leverage sequence pair information. Furthermore, as the models’ parameter sizes increase, so do their complexities, limiting their practical applicability.

Methods

In this study, we introduce a novel multi-feature independent encoding method, which encodes the gRNA-DNA sequence pair into three distinct feature matrices to minimize information loss. Additionally, we propose a lightweight hybrid deep learning framework, CRISPR-MFH, that integrates multi-scale separable convolutions and hybrid attention mechanisms for efficient and accurate off-target prediction.

Results

Extensive experiments across multiple benchmark datasets demonstrate that the proposed encoding method effectively captures critical features and that CRISPR-MFH outperforms or matches state-of-the-art models with significantly fewer parameters across multiple evaluation metrics.

Conclusions

This study offers a novel perspective for advancing deep learning technology in the realm of CRISPR-Cas9 off-target detection.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningCRISPR-Cas SystemsGene EditingRNA, Guide, CRISPR-Cas SystemsHumans

Stichwörter der Autor:innen

CRISPR-Cas9deep learninghyperparameterlightweight modeloff-target prediction

Substanzen

RNA, Guide, CRISPR-Cas Systems

Autor:innen und Institutionen

  • College of Landscape Architecture, Beijing Forestry University, Beijing 100083, China.
  • Institute of Fundamental and Frontier Sciences, University of Electronic Science and Technology of China, Chengdu 610054, China.
  • Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou 324000, China.
  • School of Mathematics and Computer Science, Zhejiang A&F University, Hangzhou 311300, China.

ORCID-Profile: Yanyi Zheng, Quan Zou, Yanpeng Yang

Förderung

  • National Natural Science Foundation of China
  • scientific research project of the education department of Zhejiang province

Zitieren

APA

Zheng, Y., Zou, Q., Li, J., & Yang, Y. (2025). CRISPR-MFH: A Lightweight Hybrid Deep Learning Framework with Multi-Feature Encoding for Improved CRISPR-Cas9 Off-Target Prediction. Genes, 16(4), 387. https://doi.org/10.3390/genes16040387

Vancouver

Zheng Y, Zou Q, Li J, Yang Y. CRISPR-MFH: A Lightweight Hybrid Deep Learning Framework with Multi-Feature Encoding for Improved CRISPR-Cas9 Off-Target Prediction. Genes. 2025;16(4):387. doi:10.3390/genes16040387

BibTeX
@article{Zheng2025,
  author = {Zheng, Yanyi and Zou, Quan and Li, Jian and Yang, Yanpeng},
  title = {CRISPR-MFH: A Lightweight Hybrid Deep Learning Framework with Multi-Feature Encoding for Improved CRISPR-Cas9 Off-Target Prediction},
  journal = {Genes},
  year = {2025},
  volume = {16},
  number = {4},
  pages = {387},
  doi = {10.3390/genes16040387}
}
Kostenlos starten →