EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

ML-ExonCNV: a robust XGBoost multi-expert ensemble framework for rare exon CNV detection in whole-exome sequencing data

Shuang-Hao Yang iD, Hua He, Shuyu Hou, Tuanfeng Yang, Zehao Yin, Hong-Yu Zhang iD, Weiyue Gu

Briefings in Bioinformatics · 2026

DOI: 10.1093/bib/bbag100 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Shuang-Hao Yang und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Bioinformatics, Chigene (Beijing) Translational Medical Research Center Co., Ltd., Beijing Yizhuang Biomedical Park, 100176, China. Hauptthemen laut MEDLINE: DNA Copy Number Variations, Exons, Exome Sequencing. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Copy number variants (CNVs) have been shown to play a significant role in the pathogenesis of various human diseases. Although several tools have been developed for detecting CNVs based on whole-exome sequencing (WES) data, their performance remains suboptimal for small exon-level CNVs (exCNVs). This is primarily due to multiple technical variabilities, including probe capture efficiency, mappability, exon size, batch effects, experimental background noise bias, and control sample selection, all of which can lead to false negatives and false positives in exCNV detection. To address these challenges, we developed ML-ExonCNV, which innovatively integrates the XGBoost machine learning model with a multi-expert ensemble approach. The model was trained using 14 features derived from 22 364 real-world, quantitative polymerase chain reaction-validated rare exCNVs. Evaluation on a test set of 492 real WES and the NA12878 gold-standard dataset demonstrated that ML-ExonCNV outperformed widely used tools such as GATK-gCNV, ExomeDepth, and CNVkit. Notably, ML-ExonCNV can detect large segmental CNVs, mosaic CNVs, and breakpoint CNVs on exon region. Furthermore, our analysis revealed recurrent exCNV-associated genes and their phenotypic correlations. Neurodevelopmental and musculoskeletal abnormalities were identified as the most frequently associated phenotypes with high-recurrence exCNVs.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

DNA Copy Number VariationsExonsExome SequencingHumansBoosting Machine Learning AlgorithmsComputational BiologyExome

Stichwörter der Autor:innen

WES CNVsXGBoostexon-level CNVsmachine learning

Autor:innen und Institutionen

  • Department of Bioinformatics, Chigene (Beijing) Translational Medical Research Center Co., Ltd., Beijing Yizhuang Biomedical Park, 100176, China.
  • Department of Bioinformatics, Beijing Quanpu Medical Laboratory Co., Ltd., E2, 3rd Floor, No. 88 Kechuang 6th Road, Beijing Yizhuang Biomedical Park, 100176, China.
  • Hubei Key Laboratory of Agricultural Bioinformatics, College of Informatics, Huazhong Agricultural University, Wuhan 430070, China.
  • Department of Neurology, Peking University International Hospital, No. 1 Life Park Road, Changping District, Beijing 102206, China.

ORCID-Profile: Shuang-Hao Yang, Hong-Yu Zhang

Zitieren

APA

Yang, S. H., He, H., Hou, S., Yang, T., Yin, Z., Zhang, H. Y., & Gu, W. (2026). ML-ExonCNV: a robust XGBoost multi-expert ensemble framework for rare exon CNV detection in whole-exome sequencing data. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag100

Vancouver

Yang SH, He H, Hou S, Yang T, Yin Z, Zhang HY, et al. ML-ExonCNV: a robust XGBoost multi-expert ensemble framework for rare exon CNV detection in whole-exome sequencing data. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag100

BibTeX
@article{Yang2026,
  author = {Yang, Shuang-Hao and He, Hua and Hou, Shuyu and Yang, Tuanfeng and Yin, Zehao and Zhang, Hong-Yu and Gu, Weiyue},
  title = {ML-ExonCNV: a robust XGBoost multi-expert ensemble framework for rare exon CNV detection in whole-exome sequencing data},
  journal = {Briefings in Bioinformatics},
  year = {2026},
  doi = {10.1093/bib/bbag100}
}
Kostenlos starten →