| Title: | PERADIGM: Phenotype embedding similarity-based rare disease gene mapping |
| Journal: | PLOS Genetics |
| Published: | 18 Dec 2025 |
| Pubmed: | https://pubmed.ncbi.nlm.nih.gov/41411243/ |
| DOI: | https://doi.org/10.1371/journal.pgen.1011976 |
Publication 17094
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
Identifying genes associated with rare diseases remains challenging due to the scarcity of patients and the limited statistical power of traditional association methods. Here, we introduce PERADIGM ( Phenotype Embedding similarity-based RAre DIsease Gene Mapping), a novel framework that leverages natural language processing techniques to integrate comprehensive phenotype information from electronic health records for rare disease gene discovery. PERADIGM employs an embedding model to capture relationships between ICD-10 codes, providing a nuanced representation of individual phenotypes. By utilizing patient similarity scores, it enhances the identification of candidate genes associated with disease-specific phenotypes, surpassing conventional methods that rely on binary disease status. We applied PERADIGM to the UK Biobank dataset for three rare diseases: autosomal dominant polycystic kidney disease (ADPKD), Marfan syndrome, and neurofibromatosis type 1 (NF1). PERADIGM identified additional candidate genes associated with ADPKD-related and Marfan syndrome-related phenotypes, some of which are supported by existing literature, and demonstrated enhanced signal detection for NF1-specific phenotypes beyond traditional methods. Our findings demonstrate the potential of PERADIGM to identify genes associated with rare diseases and related phenotypes by incorporating phenotype embeddings and patient similarity, providing a powerful tool for precision medicine and a deeper understanding of rare disease genetics and clinical manifestations.</p>
10 Keywords
- Chromosome Mapping
- Electronic Health Records
- Genetic Association Studies
- Genetic Predisposition to Disease
- Humans
- Marfan Syndrome
- Neurofibromatosis 1
- Phenotype
- Polycystic Kidney, Autosomal Dominant
- Rare Diseases
7 Authors
- Wangjie Zheng
- Yuhan Xie
- Jianlei Gu
- Hongyu Li
- Stefan Somlo
- Whitney Besse
- Hongyu Zhao
1 Application
| Application ID | Title |
|---|---|
| 29900 | Defining physical activity phenotypes based on objective measures and dissecting the genetic and phenotypic relationships of physical activity with diseases and health outcomes |
Enabling scientific discoveries that improve human health