Semantic embedding of variant effect annotations enables rapid and accurate pathogenicity prediction with VUS.Life
Jiawei Wu, Marissa Stutzman, Michael Muriello, Joy Lincoln, Donald G. Basel, Xiaowu Gai · 2026
Read Semantic embedding of variant effect annotations enables rapid and accurate pathogenicity prediction with VUS.Life by Jiawei Wu, Marissa Stutzman, Michael Muriello, Joy Lincoln, Donald G. Basel, Xiaowu Gai (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-13421479.txt.
Description
Jiawei Wu and colleagues built VUS.Life in 2026 to judge whether uncertain genetic variants cause disease, a bottleneck in genomic medicine where millions of variants lack clear meaning. The tool turns Variant Effect Predictor annotations into plain-language text, embeds them with language models, and scores new variants by closeness to known harmful ones.
Across more than 10000 variants in eight major disease genes, the method reached Matthews scores from 0.895 to 0.989 with F1 scores at or above 0.94, and its protein-level embeddings separated distinct harm mechanisms in one gene. Geneticists gain a fast, annotation-aware aid for resolving variants of uncertain significance.
Catalog record
| CREATOR | Jiawei Wu, Marissa Stutzman, Michael Muriello, Joy Lincoln, Donald G. Basel, Xiaowu Gai |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 86 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |