MCOD: a memory-constrained deep learning framework for robust outlier detection in quantitative proteomics
Jinze Huang, Huanyue Liao, Bo Meng, Guangkui Fan, Dong An, Xinhua Dai, Xiang Fang, Yang Zhao · 2026
Read MCOD: a memory-constrained deep learning framework for robust outlier detection in quantitative proteomics by Jinze Huang, Huanyue Liao, Bo Meng, Guangkui Fan, Dong An, Xinhua Dai, Xiang Fang, Yang Zhao (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13615539.txt.
Description
Jinze Huang and co-authors introduce MCOD in a 2026 paper on quality control for proteomics, the large scale study of proteins. Finding outlier samples matters, but deep learning can overfit when samples are few. The team built a method meant to stay reliable under tight memory limits.
MCOD works straight from common lab output and adds two controls. One keeps a set of prototype patterns so a few samples cannot dominate. The other adjusts sample weights as overfitting risk changes. Across simulated and real cancer data sets, it beat many rival methods and missed fewer true outliers at high precision.
Catalog record
| CREATOR | Jinze Huang, Huanyue Liao, Bo Meng, Guangkui Fan, Dong An, Xinhua Dai, Xiang Fang, Yang Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |