An Integrated Multi-Ontology Approach for Prioritizing Candidate Neurodevelopmental Disorder Genes
- 1Department of Pharmacy and Biotechnology, University of Bologna, Via Gobetti 87, 40129 Bologna, Italy
- 2Computational Genomics Platform, IRCCS University Hospital of Bologna, 40138 Bologna, Italy
Abstract
Rare neurodevelopmental disorders (NDDs) are clinically and genetically heterogeneous conditions affecting brain function, and a comprehensive set of NDD-associated genes is still lacking. We integrated gene lists from Orphanet, SFARI Gene, GeneTrek, and a large-scale genomic study, and classified 19,354 human protein-coding genes into 2639 high-confidence genes, 6856 candidate genes, and 9859 genes with no NDD evidence. Enrichment analysis of the high-confidence set highlighted nervous system development, synaptic processes, tRNA aminoacylation, and cancer-related pathways, and Monarch Disease Ontology (MONDO) enrichment confirmed its association with NDDs. We then prioritized candidate genes with a Multi-Ontology Enrichment (MOE) score, which counts the annotation resources in which a gene shares enriched terms with the high-confidence set. Higher MOE scores were associated with stronger loss-of-function constraint in the Genome Aggregation Database (gnomAD) a larger fraction of genes with pathogenic ClinVar variants (10.7% to 44.0%), and more frequent gene-level association in NDD sequencing cohorts (2.8% to 8.3%), and SysNDD scores provided further support. Part of this signal reflects how well genes are annotated, which we discuss as a limitation. Overall, this work provides a harmonized NDD gene resource and a transparent prioritization of candidate genes for diagnostic and functional studies.
Related
- Prioritizing Sarcopenia-Related Genes from Cross-Trait Candidate Regions: Regulatory Variant Mapping and Phenome-Wide Mendelian Randomization
- MED20 biallelic pathogenic variants cause a neurodevelopmental disorder altering both transcription activity and Transcription-Coupled Repair pathway
- 5q31 duplications encompassing PURA are associated with a neurodevelopmental disorder
- Integrated network toxicology and experimental validation identify VCAM1 as a candidate molecular link between pyrethroid exposure and keloid
- Integrated multi-omics analysis of flavonoid metabolism in soybean seeds of different colours reveals candidate regulatory networks associated with anthocyanin and isoflavone accumulation
BibTeX
@article{rivi2026integrated,
title = {{An Integrated Multi-Ontology Approach for Prioritizing Candidate Neurodevelopmental Disorder Genes}},
author = {Chiara Rivi and Maria Paola Turina and Emidio Capriotti},
journal = {International Journal of Molecular Sciences},
year = {2026},
month = nov,
doi = {10.3390/ijms272210272},
url = {https://www.mdpi.com/1422-0067/27/22/10272},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - An Integrated Multi-Ontology Approach for Prioritizing Candidate Neurodevelopmental Disorder Genes
AU - Chiara Rivi
AU - Maria Paola Turina
AU - Emidio Capriotti
PY - 2026
DA - 2026/11/16
JO - International Journal of Molecular Sciences
DO - 10.3390/ijms272210272
UR - https://www.mdpi.com/1422-0067/27/22/10272
AB - Rare neurodevelopmental disorders (NDDs) are clinically and genetically heterogeneous conditions affecting brain function, and a comprehensive set of NDD-associated genes is still lacking. We integrated gene lists from Orphanet, SFARI Gene, GeneTrek, and a large-scale genomic study, and classified 19,354 human protein-coding genes into 2639 high-confidence genes, 6856 candidate genes, and 9859 genes with no NDD evidence. Enrichment analysis of the high-confidence set highlighted nervous system development, synaptic processes, tRNA aminoacylation, and cancer-related pathways, and Monarch Disease Ontology (MONDO) enrichment confirmed its association with NDDs. We then prioritized candidate genes with a Multi-Ontology Enrichment (MOE) score, which counts the annotation resources in which a gene shares enriched terms with the high-confidence set. Higher MOE scores were associated with stronger loss-of-function constraint in the Genome Aggregation Database (gnomAD) a larger fraction of genes with pathogenic ClinVar variants (10.7% to 44.0%), and more frequent gene-level association in NDD sequencing cohorts (2.8% to 8.3%), and SysNDD scores provided further support. Part of this signal reflects how well genes are annotated, which we discuss as a limitation. Overall, this work provides a harmonized NDD gene resource and a transparent prioritization of candidate genes for diagnostic and functional studies.
SN - 1422-0067
ER -