AutoPM3: enhancing variant interpretation via LLM-driven PM3 evidence extraction from scientific literature
Bioinformatics · 01.07.2025 · 41 (7)
DOI: 10.1093/bioinformatics/btaf382 ↗Kurz gefasst
Fachartikel in der Zeitschrift Bioinformatics, online erschienen am 30.06.2025. Verfasst von Shumin Li und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, School of Computing and Data Science, University of Hong Kong, Hong Kong, 999077, China. Hauptthemen laut MEDLINE: Genetic Variation, Software, Data Mining, Computational Biology. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hong Kong Research Grants Council grants General Research Fund, Shenzhen Municipal Government General Program, University Research Committee fund from HKU.
Abstract (Originalsprache)
Motivation Rare diseases affect over 300 million people worldwide and are often caused by genetic variants. While variant detection has become cost-effective, interpreting these variants—particularly collecting literature-based evidence like ACMG/AMP PM3—remains complex and time-consuming. Results We present AutoPM3, a method that automates PM3 evidence extraction from literatures using open-source large language models (LLMs). AutoPM3 combines a Text2SQL-based variant extractor and a retrieval-augmented generation (RAG) module, enhanced by a variant-specific retriever and fine-tuned LLM, to separately process tables and text. We curated PM3-Bench, a dataset of 1027 variant-publication evidence pairs from ClinGen. On openly accessible pairs, AutoPM3 achieved 86.1% accuracy for variant hits and 72.5% recall for in trans variants—outperforming other methods, including those using larger models. We uncovered the effectiveness of AutoPM3’s key modules, especially for variant-specific retriever and Text2SQL, through the sequential ablation study. AutoPM3 located evidence in 76 s, demonstrating that open-source LLMs can offer an efficient, cost-effective solution for rare disease diagnosis. Availability and implementation AutoPM3 is implemented and freely available under the MIT license at https://github.com/HKU-BAL/AutoPM3.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Computer Science, School of Computing and Data Science, University of Hong Kong, Hong Kong, 999077, China.
- School of Biomedical Sciences, University of Hong Kong, Hong Kong, 999077, China.
- Hong Kong Genome Institute, Hong Kong, 999077, China.
- Department of Statistics and Actuarial Science, The University of Hong Kong, Hong Kong, 999077, China.
- Center for Translational Stem Cell Biology, Hong Kong Science and Technology Park, Hong Kong, 999077, China.
ORCID-Profile: Shumin Li, Ruibang Luo
Förderung
- Hong Kong Research Grants Council grants General Research Fund
- Shenzhen Municipal Government General Program
- University Research Committee fund from HKU
Zitieren
APA
Li, S., Wang, Y., Liu, C. M., Huang, Y., Lam, T. W., & Luo, R. (2025). AutoPM3: enhancing variant interpretation via LLM-driven PM3 evidence extraction from scientific literature. Bioinformatics, 41(7). https://doi.org/10.1093/bioinformatics/btaf382
Vancouver
Li S, Wang Y, Liu CM, Huang Y, Lam TW, Luo R. AutoPM3: enhancing variant interpretation via LLM-driven PM3 evidence extraction from scientific literature. Bioinformatics. 2025;41(7). doi:10.1093/bioinformatics/btaf382
BibTeX
@article{Li2025,
author = {Li, Shumin and Wang, Yiding and Liu, Chi-Man and Huang, Yuanhua and Lam, Tak-Wah and Luo, Ruibang},
title = {AutoPM3: enhancing variant interpretation via LLM-driven PM3 evidence extraction from scientific literature},
journal = {Bioinformatics},
year = {2025},
volume = {41},
number = {7},
doi = {10.1093/bioinformatics/btaf382}
}
EUVIMED