EUVIMEDEUVIMED entdecken
Systematische ÜbersichtOpen Access · CC BY 4.0

Deep learning approaches for resolving genomic discrepancies in cancer: a systematic review and clinical perspective

Muhammad Zubair iD, Ali Haider Khan, Syed Fakhar Bilal, Jianqiang Li

Briefings in Bioinformatics · 01.11.2025 · 26 (6)

DOI: 10.1093/bib/bbaf541 ↗

Kurz gefasst

Systematische Übersicht in der Zeitschrift Briefings in Bioinformatics, online erschienen am 02.11.2025. Verfasst von Muhammad Zubair und 3 weiteren Autor:innen. Beteiligte Einrichtung: Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China. Hauptthemen laut MEDLINE: Deep Learning, Neoplasms, Genomics. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 103 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Discrepancies in cancer sequencing data continue to pose significant challenges for accurate mutation detection, potentially resulting in misdiagnoses and suboptimal treatment strategies. Although deep learning (DL) has emerged as a transformative approach for identifying and rectifying these errors, there remains a lack of comprehensive evaluation of DL architectures, performance benchmarks, and clinical translation. In this systematic review of 78 studies (2015–2024), We synthesize recent advancements in DL methodologies for identifying genomic discrepancies, demonstrating that convolutional and graph-based architectures currently achieve state-of-the-art performance in variant calling and tumor stratification. DL models reduce false-negative rates by 30%–40% compared to traditional pipelines, with methods such as MAGPIE prioritizing pathogenic variants with 92% accuracy. However, challenges such as data scarcity, batch effects, and the interpretability of “black-box” models persist. We propose a future research roadmap advocating federated learning to enhance data privacy and attention mechanisms to improve model transparency. By bridging bioinformatics and oncology, this review offers actionable insights to expedite the deployment of DL in precision cancer therapy.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningNeoplasmsGenomicsHumansComputational BiologyMutation

Stichwörter der Autor:innen

cancer genomicsclinical bioinformaticsdeep learninggenomic discrepanciesmulti-omics integrationmutation detectionvariant callingwhole exome sequencing (WES)

Autor:innen und Institutionen

  • Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China.

ORCID-Profile: Muhammad Zubair

Zitieren

APA

Zubair, M., Khan, A. H., Bilal, S. F., & Li, J. (2025). Deep learning approaches for resolving genomic discrepancies in cancer: a systematic review and clinical perspective. Briefings in Bioinformatics, 26(6). https://doi.org/10.1093/bib/bbaf541

Vancouver

Zubair M, Khan AH, Bilal SF, Li J. Deep learning approaches for resolving genomic discrepancies in cancer: a systematic review and clinical perspective. Briefings in Bioinformatics. 2025;26(6). doi:10.1093/bib/bbaf541

BibTeX
@article{Zubair2025,
  author = {Zubair, Muhammad and Khan, Ali Haider and Bilal, Syed Fakhar and Li, Jianqiang},
  title = {Deep learning approaches for resolving genomic discrepancies in cancer: a systematic review and clinical perspective},
  journal = {Briefings in Bioinformatics},
  year = {2025},
  volume = {26},
  number = {6},
  doi = {10.1093/bib/bbaf541}
}
Kostenlos starten →