Back

EAGLE-AI: A large language model workflow for automated extraction and scoring of literature evidence linking genes to autism spectrum disorder

Furlan, V.; Moran, J.; Salazar, N. B.; Rennie, O.; Hoang, N.; wan, A.; Mendes de Aquino, M.; Engchuan, W.; Vorstman, J. A. S.; Scherer, S. W.

2025-10-02 genetic and genomic medicine
10.1101/2025.09.10.25334730 medRxiv
Show abstract

We previously developed the Evaluation of Autism Gene Link Evidence (EAGLE) manual curation framework and used it to characterise 219 autism-associated genes. However, this effort took years of human work. We present EAGLE-AI, an automated evidence collection, screening, extraction, and scoring system incorporating agentic large language model (LLM) workforces. On a test set of 116 manuscripts screened for ease of machine-readability, EAGLE-AI achieves F1 score of 91% in reproducing human curators data extractions. Its evidence scores differ from those of human scorers by 14.3%. Our findings indicate that EAGLE-AI can successfully automate most of a clinical genomics evidence curation process.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.