A multi-label cascade flexible neural forest model for predicting the subcellular location of multi-site bacterial proteins
Lianxin Zhong, Yang Li, Yanhui Cheng, Guangchang Han, Anqi Li · 2026
Read A multi-label cascade flexible neural forest model for predicting the subcellular location of multi-site bacterial proteins by Lianxin Zhong, Yang Li, Yanhui Cheng, Guangchang Han, Anqi Li (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/PMC13575143.txt.
Description
Lianxin Zhong and four colleagues propose a model that predicts where bacterial proteins end up inside the cell, in a 2026 paper. Some proteins work at more than one site, which makes the task a multi-label problem. Links between labels, small datasets, and low sequence similarity all trip up standard methods.
Their model, named MLCFN Forest, codes label sets together, then sorts samples through layers of flexible neural trees. Hard cases get extra model capacity as they move through the layers. On standard Gram-negative and Gram-positive datasets it beat five rival methods, reaching about 78 to 81 percent on its main accuracy scores. It also held up after features were heavily compressed. The authors note that outside test sets are still lacking.
Catalog record
| CREATOR | Lianxin Zhong, Yang Li, Yanhui Cheng, Guangchang Han, Anqi Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43811 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |