PhenoDis: a comprehensive database for phenotypic characterization of rare cardiac diseases

Angela Adler,
Pia Kirchmeier,
Julian Reinhard,
Barbara Brauner,
Irmtraud Dunger,
Gisela Fobo,
Goar Frishman,
Corinna Montrone,
H.-Werner Mewes,
Matthias Arnold and
Andreas RueppEmail author View ORCID ID profile

Orphanet Journal of Rare Diseases201813:22

https://doi.org/10.1186/s13023-018-0765-y

Received: 11 September 2017

Accepted: 12 January 2018

Published: 25 January 2018

Abstract

Background

Thoroughly annotated data resources are a key requirement in phenotype dependent analysis and diagnosis of diseases in the area of precision medicine. Recent work has shown that curation and systematic annotation of human phenome data can significantly improve the quality and selectivity for the interpretation of inherited diseases. We have therefore developed PhenoDis, a comprehensive, manually annotated database providing symptomatic, genetic and imprinting information about rare cardiac diseases.

Results

PhenoDis includes 214 rare cardiac diseases from Orphanet and 94 more from OMIM. For phenotypic characterization of the diseases, we performed manual annotation of diseases with articles from the biomedical literature. Detailed description of disease symptoms required the use of 2247 different terms from the Human Phenotype Ontology (HPO). Diseases listed in PhenoDis frequently cover a broad spectrum of symptoms with 28% from the branch of ‘cardiovascular abnormality’ and others from areas such as neurological (11.5%) and metabolism (6%). We collected extensive information on the frequency of symptoms in respective diseases as well as on disease-associated genes and imprinting data. The analysis of the abundance of symptoms in patient studies revealed that most of the annotated symptoms (71%) are found in less than half of the patients of a particular disease. Comprehensive and systematic characterization of symptoms including their frequency is a pivotal prerequisite for computer based prediction of diseases and disease causing genetic variants. To this end, PhenoDis provides in-depth annotation for a complete group of rare diseases, including information on pathogenic and likely pathogenic genetic variants for 206 diseases as listed in ClinVar. We integrated all results in an online database (http://mips.helmholtz-muenchen.de/phenodis/) with multiple search options and provide the complete dataset for download.

Conclusion

PhenoDis provides a comprehensive set of manually annotated rare cardiac diseases that enables computational approaches for disease prediction via decision support systems and phenotype-driven strategies for the identification of disease causing genes.

Keywords

Rare cardiac diseasesHeartPhenotypeGenotypePrecision medicineGenetic disordersDecision support systemsBioinformaticsMedical geneticsPersonalized medicine