LitGene transformer integrates text and ontology to improve gene representations
Researchers introduced LitGene, an interpretable transformer that blends unstructured biomedical literature with Gene Ontology annotations using contrastive learning. The model, released as open‑source code with a public web interface, was evaluated on eight benchmark tasks for protein properties and achieved competitive accuracy while also supporting zero‑shot prediction of GO terms, KEGG…
Key points
- LitGene combines literature text and Gene Ontology via contrastive learning in an interpretable transformer.
- It attains competitive accuracy on eight protein property benchmarks and can zero‑shot predict GO, KEGG, and disease associations.
- The model identified ANGPTL6 as a candidate gene for idiopathic pulmonary fibrosis, corroborated by independent expression data.
In a case study, LitGene highlighted ANGPTL6 as a potential candidate gene for idiopathic pulmonary fibrosis (IPF). Independent IPF samples showed differential expression of this gene, and domain experts reviewed the prediction as biologically plausible. A SHAP‑based interpretability layer points to specific text features driving each prediction, and an automated literature‑grounding system supplies supporting evidence when available. The work was funded by the Ovarian Cancer Research Alliance and two NIH grants, and the authors declared no competing interests.
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Research
All →- Anthropic’s Claude Mythos 5 redesigns quantum circuits better than prior research · 1 src
- Google DeepMind tests Gemini 4 Argon on Apollo 2 humanoid in internal test · 1 src
- MIT study finds AI solved 35% of theory problems by mid-2026 · 4 src
- Former OpenAI researcher says new theory may explain AI thinking, could affect safety · 2 src
- Author builds CodeOwnership tool to check AI-generated code ownership · 1 src
Comments
via GitHub Discussions