MCOD: a memory-constrained deep learning framework for robust outlier detection in quantitative proteomics
Briefings in Bioinformatics · 2026
DOI: 10.1093/bib/bbag507 ↗Kurz gefasst
Fachartikel in der Zeitschrift Briefings in Bioinformatics, 2026. Verfasst von Jinze Huang und 7 weiteren Autor:innen. Beteiligte Einrichtung: Technology Innovation Center of Mass Spectrometry for State Market Regulation, Center for Advanced Measurement Science, National Institute of Metrolo…. Hauptthemen laut MEDLINE: Deep Learning, Proteomics. Die Arbeit nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Research Project of the National Institute of Metrology, National Natural Science Foundation of China, Plan for Leading Talents of Science and Technology Innovation.
Abstract (Originalsprache)
Ensuring robust quality control (QC) remains a major challenge in quantitative proteomics, particularly in detecting and managing outliers. Deep learning offers powerful representational capacity for ultra-high-dimensional data but often suffers from overfitting in small-sample scenarios. To address this, we propose memory-constrained outlier detection (MCOD), a deep anomaly detection framework that directly processes MaxQuant outputs and achieves competitive recall at high precision levels, which suggests a reduced risk of overlooking true outliers. MCOD integrates two innovations: (i) a memory-constrained module (MC module) that mitigates over-representation of samples via prototype-based regularization, and (ii) an adaptive steady-aware regulator that dynamically adjusts the per-sample loss weights in the MC module according to the estimated overfitting risk. Across two simulation settings based on a human cervical cancer cell line (HeLa) proteomics dataset and two real-world cancer proteomics datasets, MCOD consistently outperformed 18 statistical, machine learning, and deep learning baselines, achieving superior area under the receiver operating characteristic curve and area under the precision-recall curve scores. Functional enrichment analyses on the two real-world datasets showed that MCOD performed favorably compared to the three domain-specific models. Furthermore, feature-level visualization provided insights into the rationale behind the model’s anomaly assignments. Collectively, MCOD establishes a robust and scalable framework for data QC in quantitative proteomics.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Technology Innovation Center of Mass Spectrometry for State Market Regulation, Center for Advanced Measurement Science, National Institute of Metrology, 18, Beisanhuandonglu, Chaoyang District, 100029, Beijing, China.
- College of Information and Electrical Engineering, China Agricultural University, No. 17 Tsinghua East Road, Haidian District, Beijing 100083, China.
- China National Institute of Standardization, No. 9 Madian East Road, Haidian District, Beijing 100191, China.
ORCID-Profile: Jinze Huang, Yang Zhao
Förderung
- Research Project of the National Institute of Metrology
- National Natural Science Foundation of China
- Plan for Leading Talents of Science and Technology Innovation
- National Key R&D Program of China
Zitieren
APA
Huang, J., Liao, H., Meng, B., Fan, G., An, D., Dai, X., Fang, X., & Zhao, Y. (2026). MCOD: a memory-constrained deep learning framework for robust outlier detection in quantitative proteomics. Briefings in Bioinformatics. https://doi.org/10.1093/bib/bbag507
Vancouver
Huang J, Liao H, Meng B, Fan G, An D, Dai X, et al. MCOD: a memory-constrained deep learning framework for robust outlier detection in quantitative proteomics. Briefings in Bioinformatics. 2026. doi:10.1093/bib/bbag507
BibTeX
@article{Huang2026,
author = {Huang, Jinze and Liao, Huanyue and Meng, Bo and Fan, Guangkui and An, Dong and Dai, Xinhua and Fang, Xiang and Zhao, Yang},
title = {MCOD: a memory-constrained deep learning framework for robust outlier detection in quantitative proteomics},
journal = {Briefings in Bioinformatics},
year = {2026},
doi = {10.1093/bib/bbag507}
}
EUVIMED