PhenoLinker: Phenotype-Gene Link Prediction and Explanation using Heterogeneous Graph Neural Networks
Authors:
Preprint
Abstract
The association of a given human phenotype to a genetic variant remains a critical challenge for biology. We present a novel system called PhenoLinker capable of associating a score to a phenotype-gene relationship by using heterogeneous information networks and a convolutional neural network-based model for graphs, which can provide an explanation for the predictions. This system can aid in the discovery of new associations and in the understanding of the consequences of human genetic variation.Keywords
All Related Versions
This publication has 49 references indexed in Scilit:
- A novel de novo pathogenic variant in KDM3B gene at the first Albanian case of Diets-Jongmans syndrome: DIJOSMolecular Genetics and Metabolism Reports, 2022
- Graph neural network approaches for drug-target interactionsCurrent Opinion in Structural Biology, 2022
- Autosomal recessive variants in TUBGCP2 alter the γ-tubulin ring complex leading to neurodevelopmental diseaseiScience, 2021
- Bi-allelic Pathogenic Variants in TUBGCP2 Cause Microcephaly and Lissencephaly Spectrum DisordersThe American Journal of Human Genetics, 2019
- UCP2 gene polymorphisms in obesity and diabetes, and the role of UCP2 in cancerFEBS Letters, 2019
- HPO2Vec+: Leveraging heterogeneous knowledge resources to enrich node embeddings for the Human Phenotype OntologyJournal of Biomedical Informatics, 2019
- De Novo and Inherited Pathogenic Variants in KDM3B Cause Intellectual Disability, Short Stature, and Facial DysmorphismThe American Journal of Human Genetics, 2019
- A Ribosomopathy Reveals Decoding Defective Ribosomes Driving Human DysmorphismThe American Journal of Human Genetics, 2017
- Association of the Charcot–Marie–Tooth disease gene ARHGEF10 with paclitaxel induced peripheral neuropathy in NCCTG N08CA (Alliance)Journal of the Neurological Sciences, 2015
- Gene Ontology: tool for the unification of biologyNature Genetics, 2000