scLTF: A self-training clustering framework for large-scale single-cell RNA-Seq data
Hongyi Yuan, Chunyan Wang, Qiucheng Sun, Mingyu Sun, Mujie Fan · 2026
Read scLTF: A self-training clustering framework for large-scale single-cell RNA-Seq data by Hongyi Yuan, Chunyan Wang, Qiucheng Sun, Mingyu Sun, Mujie Fan (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/plos-pone-0359153.txt.
Description
Hongyi Yuan and four colleagues present a software method, published in PLOS ONE in 2026, for sorting huge collections of single cells by their gene activity. Modern sequencing can profile hundreds of thousands of cells at once, but grouping those cells into types gets slow and unreliable at that scale. Their framework, called scLTF, is built to handle the volume.
The method first makes rough groups with a fast graph based algorithm, then picks the most typical cells from each group. A transformer model, the kind of network behind modern language tools, then refines the groups step by step. On several real datasets it matched or beat leading deep learning methods on accuracy while using only a fraction of their running time.
Catalog record
| CREATOR | Hongyi Yuan, Chunyan Wang, Qiucheng Sun, Mingyu Sun, Mujie Fan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 76107 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |