GenPhenia: using deep neural networks to accelerate rare-disease diagnosis
- 1Departamento de Química Biológica, Facultad de Ciencias Exactas y Naturales, Universidad de Buenos Aires (FCEyN-UBA) e Instituto de Química Biológica de la Facultad de Ciencias Exactas y Naturales (IQUIBICEN), CONICET, Pabellón 2 de Ciudad Universitaria, C1428EHA Ciudad de Buenos Aires, Argentina
- 2Análisis de Datos Genómicos, Camino Parque Centenario Nº 2565 - La Plata, Alicia Moreau de Justo Nº 1750 3º H - CABA, Buenos Aires, Argentina
Abstract
Background: Traditional molecular diagnosis of rare genetic diseases often struggles with complex genotype–phenotype relationships, such as pleiotropy and locus heterogeneity. Despite advances in next-generation sequencing, variant interpretation remains a major bottleneck, frequently requiring labor-intensive expert review to establish phenotypic specificity. To address this challenge, we developed GenPhenia, a deep learning–based framework designed to assist phenotype-driven gene prioritization using clinical features encoded with the Human Phenotype Ontology (HPO).
Results: GenPhenia leverages large-scale synthetic clinical history datasets and a graph neural network architecture with message-passing strategies combined with BERT-based phenotype embeddings to learn phenotype–gene associations. The model was validated using both ClinVar-derived datasets and real-world clinical cohorts. On the challenging MCRD benchmark dataset, GenPhenia achieved 60% Top-1 recall and 80% Top-10 recall, substantially outperforming previously reported methods, which achieved less than 30% Top-10 recall under the same conditions. Clinical case studies across endocrinology, cardiology, and immunology further demonstrate that the model effectively refines gene rankings as phenotypic specificity increases during the diagnostic process.
Conclusions: GenPhenia provides a scalable and highly sensitive computational approach for phenotype-driven gene prioritization that aligns with the iterative nature of clinical phenotyping and genomic diagnosis. The framework may facilitate more efficient variant interpretation and support the integration of artificial intelligence methods into clinical genomics workflows for rare disease diagnosis.
Graphical Abstract:
Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s40246-026-01023-9.
Related
- Neural networks as decision trees: an analytical framework for learning and neural selectivity
- Structural Evolution of Neural Networks during Zebrafish Early Development
- Shaping Slow Manifolds in Equivariant Recurrent Neural Networks
- Protein Language Model-Conditioned Graph Neural Networks for Multitask GPCR Ligand Activity Prediction
- Distributed implantable microchips as wireless networks for neural sensing and stimulation
BibTeX
@article{brunello2026genphenia,
title = {{GenPhenia: using deep neural networks to accelerate rare-disease diagnosis}},
author = {Franco Gino Brunello and Gianlucca Colangelo and Ana Rius and Lorenzo Erra and Ana Clara Lugones and Nicolás Aguirre and Germán Biagioli and Jonathan Zaiat and Adrián Turjanski and Marcelo A. Martí},
journal = {Human Genomics},
year = {2026},
month = aug,
doi = {10.1186/s40246-026-01023-9},
url = {https://doi.org/10.1186/s40246-026-01023-9},
note = {CC BY-NC-ND 4.0},
}
RIS
TY - JOUR
TI - GenPhenia: using deep neural networks to accelerate rare-disease diagnosis
AU - Franco Gino Brunello
AU - Gianlucca Colangelo
AU - Ana Rius
AU - Lorenzo Erra
AU - Ana Clara Lugones
AU - Nicolás Aguirre
AU - Germán Biagioli
AU - Jonathan Zaiat
AU - Adrián Turjanski
AU - Marcelo A. Martí
PY - 2026
DA - 2026/08/14
JO - Human Genomics
DO - 10.1186/s40246-026-01023-9
UR - https://doi.org/10.1186/s40246-026-01023-9
AB - Background: Traditional molecular diagnosis of rare genetic diseases often struggles with complex genotype–phenotype relationships, such as pleiotropy and locus heterogeneity. Despite advances in next-generation sequencing, variant interpretation remains a major bottleneck, frequently requiring labor-intensive expert review to establish phenotypic specificity. To address this challenge, we developed GenPhenia, a deep learning–based framework designed to assist phenotype-driven gene prioritization using clinical features encoded with the Human Phenotype Ontology (HPO). Results: GenPhenia leverages large-scale synthetic clinical history datasets and a graph neural network architecture with message-passing strategies combined with BERT-based phenotype embeddings to learn phenotype–gene associations. The model was validated using both ClinVar-derived datasets and real-world clinical cohorts. On the challenging MCRD benchmark dataset, GenPhenia achieved 60% Top-1 recall and 80% Top-10 recall, substantially outperforming previously reported methods, which achieved less than 30% Top-10 recall under the same conditions. Clinical case studies across endocrinology, cardiology, and immunology further demonstrate that the model effectively refines gene rankings as phenotypic specificity increases during the diagnostic process. Conclusions: GenPhenia provides a scalable and highly sensitive computational approach for phenotype-driven gene prioritization that aligns with the iterative nature of clinical phenotyping and genomic diagnosis. The framework may facilitate more efficient variant interpretation and support the integration of artificial intelligence methods into clinical genomics workflows for rare disease diagnosis. Graphical Abstract: Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s40246-026-01023-9.
SN - 1473-9542
SN - 1479-7364
ER -