PhenoSS: phenotype semantic similarity-based approach for rare disease prediction and patient clustering

Shihan Chen; Quan M Nguyen; Yu Hu; Cong Liu; Chunhua Weng; Kai Wang

doi:10.1093/jamia/ocag131

July 14, 2026Journal of the American Medical Informatics Association

PhenoSS: phenotype semantic similarity-based approach for rare disease prediction and patient clustering

View Full Paper

Ask AI

Bookmark

Share

Key Points

To develop a scalable, statistically principled framework that accounts for term dependencies, hierarchical structures, and documentation batch effects in rare disease phenotyping and diagnosis.
Developed PhenoSS, a Gaussian copula framework modeling disease-specific Human Phenotype Ontology (HPO) term prevalence and joint dependencies via multivariate normal distributions.
Integrated phenotype frequencies from external databases (OARD and HPO) alongside a dedicated batch-effect correction module.
Evaluated patient clustering and posterior odds-based disease prioritization across simulated benchmarks and real-world electronic health record datasets.
Simulations demonstrated robust disease-prediction capability with consistent accuracy gains following batch-effect correction across diverse data collection environments.
Electronic health record evaluations confirmed PhenoSS identified clinically coherent patient clusters and reliably separated distinct rare disease profiles.
Disease prioritization performance matched or surpassed existing tools, exhibiting particular resilience when handling sparse or noisy phenotype annotations.

No abstract yet

We only have this paper's title and publication details.

Read it at the source

Implication

Computational modeling study demonstrates robust disease prioritization and stratification in rare disease cohorts, indicating improved diagnostic accuracy despite noisy or sparse clinical records.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Chen et al. (2026) studied this question.

synapsesocial.com/papers/6a708721660549caf2c520fb https://doi.org/10.1093/jamia/ocag131

View Full Paper

Ask AI

Bookmark

Share

PhenoSS: phenotype semantic similarity-based approach for rare disease prediction and patient clustering

Key Points

No abstract yet

Implication

Discussion

Cite This Study

PhenoSS: phenotype semantic similarity-based approach for rare disease prediction and patient clustering

Key Points

No abstract yet

Implication

Discussion

Cite This Study

Keep reading

Keep reading