Pathogenicity prediction for noncanonical splice-altering variants based on multimodal feature fusion
Xiaoyan Li, Zhen Peng, Yiran Zhao, Shuhan Wang, Xingpeng Zhou, Xiongjian Luo, Yansen Su, Chunhou Zheng, Junfeng Xia · 2026
Read Pathogenicity prediction for noncanonical splice-altering variants based on multimodal feature fusion by Xiaoyan Li, Zhen Peng, Yiran Zhao, Shuhan Wang, Xingpeng Zhou, Xiongjian Luo, Yansen Su, Chunhou Zheng, Junfeng Xia (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-13365626.txt.
Description
Xiaoyan Li and co-authors build a predictor for harmful gene variants that alter splicing outside the usual sites, in work from 2026. Splicing errors are the second most common class of disease-causing variants. Tools judge the classic splice sites well but struggle elsewhere in the gene.
Their model fuses three signal types into one score. A pretrained DNA model supplies long-range context around each variant. Local pattern nets add fine sequence detail, and known gene facts add a third view. A gated mixer weighs the views case by case. Tests on several held-out sets beat top rival tools, including a widely used scoring system.
Catalog record
| CREATOR | Xiaoyan Li, Zhen Peng, Yiran Zhao, Shuhan Wang, Xingpeng Zhou, Xiongjian Luo, Yansen Su, Chunhou Zheng, Junfeng Xia |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |