PhenoScore quantifies phenotypic variation for rare genetic diseases by combining facial analysis with other clinical features using a machine-learning framework

Alexander J M Dingemans; Max Hinne; Kim M G Truijen; Lia Goltstein; Jeroen van Reeuwijk; Nicole de Leeuw; Janneke Schuurs-Hoeijmakers; Rolph Pfundt; Illja J Diets; Joery den Hoed; Elke de Boer; Jet Coenen-van der Spek; Sandra Jansen; Bregje W van Bon; Noraly Jonis; Charlotte W Ockeloen; Anneke T Vulto-van Silfhout; Tjitske Kleefstra; David A Koolen; Philippe M Campeau; Elizabeth E Palmer; Hilde Van Esch; Gholson J Lyon; Fowzan S Alkuraya; Anita Rauch; Ronit Marom; Diana Baralle; Pleuntje J van der Sluijs; Gijs W E Santen; R Frank Kooy; Marcel A J van Gerven; Lisenka E L M Vissers; Bert B A de Vries

doi:10.1038/s41588-023-01469-w

PhenoScore quantifies phenotypic variation for rare genetic diseases by combining facial analysis with other clinical features using a machine-learning framework

Nat Genet. 2023 Sep;55(9):1598-1607. doi: 10.1038/s41588-023-01469-w. Epub 2023 Aug 7.

Authors

Alexander J M Dingemans^{1

2}, Max Hinne², Kim M G Truijen¹, Lia Goltstein¹, Jeroen van Reeuwijk¹, Nicole de Leeuw¹, Janneke Schuurs-Hoeijmakers¹, Rolph Pfundt¹, Illja J Diets¹, Joery den Hoed³, Elke de Boer¹, Jet Coenen-van der Spek¹, Sandra Jansen⁴, Bregje W van Bon¹, Noraly Jonis¹, Charlotte W Ockeloen¹, Anneke T Vulto-van Silfhout¹, Tjitske Kleefstra¹, David A Koolen¹, Philippe M Campeau⁵, Elizabeth E Palmer^{6

7}, Hilde Van Esch⁸, Gholson J Lyon^{9

10}, Fowzan S Alkuraya¹¹, Anita Rauch¹², Ronit Marom¹³, Diana Baralle¹⁴, Pleuntje J van der Sluijs¹⁵, Gijs W E Santen¹⁵, R Frank Kooy¹⁶, Marcel A J van Gerven², Lisenka E L M Vissers^#¹⁷, Bert B A de Vries^#¹⁸

Affiliations

¹ Department of Human Genetics, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Center, Nijmegen, the Netherlands.
² Department of Artificial Intelligence, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands.
³ Language and Genetics Department, Max Planck Institute for Psycholinguistics, Nijmegen, the Netherlands.
⁴ Department of Human Genetics, Amsterdam UMC, University of Amsterdam, Amsterdam, the Netherlands.
⁵ Department of Pediatrics, University of Montreal, Montreal, Quebec, Canada.
⁶ Faculty of Medicine and Health, UNSW Sydney, Sydney, New South Wales, Australia.
⁷ Sydney Children's Hospitals Network, Sydney, New South Wales, Australia.
⁸ Center for Human Genetics, University Hospitals Leuven, University of Leuven, Leuven, Belgium.
⁹ Department of Human Genetics and George A. Jervis Clinic, Institute for Basic Research in Developmental Disabilities (IBR), Staten Island, NY, USA.
¹⁰ Biology PhD Program, The Graduate Center, The City University of New York, New York City, NY, USA.
¹¹ Department of Translational Genomics, Center for Genomic Medicine, King Faisal Specialist Hospital and Research Center, Riyadh, Saudi Arabia.
¹² Institute of Medical Genetics, University of Zürich, Zürich, Switzerland.
¹³ Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA.
¹⁴ Faculty of Medicine, University of Southampton, Southampton, UK.
¹⁵ Department of Clinical Genetics, Leiden University Medical Center, Leiden, the Netherlands.
¹⁶ Department of Medical Genetics, University of Antwerp, Antwerp, Belgium.
¹⁷ Department of Human Genetics, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Center, Nijmegen, the Netherlands. [email protected].
¹⁸ Department of Human Genetics, Donders Institute for Brain, Cognition and Behaviour, Radboud University Medical Center, Nijmegen, the Netherlands. [email protected].

^# Contributed equally.

Abstract

Several molecular and phenotypic algorithms exist that establish genotype-phenotype correlations, including facial recognition tools. However, no unified framework that investigates both facial data and other phenotypic data directly from individuals exists. We developed PhenoScore: an open-source, artificial intelligence-based phenomics framework, combining facial recognition technology with Human Phenotype Ontology data analysis to quantify phenotypic similarity. Here we show PhenoScore's ability to recognize distinct phenotypic entities by establishing recognizable phenotypes for 37 of 40 investigated syndromes against clinical features observed in individuals with other neurodevelopmental disorders and show it is an improvement on existing approaches. PhenoScore provides predictions for individuals with variants of unknown significance and enables sophisticated genotype-phenotype studies by testing hypotheses on possible phenotypic (sub)groups. PhenoScore confirmed previously known phenotypic subgroups caused by variants in the same gene for SATB1, SETBP1 and DEAF1 and provides objective clinical evidence for two distinct ADNP-related phenotypes, already established functionally.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Algorithms
Artificial Intelligence*
Biological Variation, Population
DNA-Binding Proteins
Humans
Machine Learning
Matrix Attachment Region Binding Proteins*
Phenotype
Transcription Factors

Substances

SATB1 protein, human
Matrix Attachment Region Binding Proteins
DEAF1 protein, human
DNA-Binding Proteins
Transcription Factors

Grants and funding

R35 GM133408/GM/NIGMS NIH HHS/United States