Leveraging Next-Generation Phenotyping in Dysmorphology to Support Variant Interpretation in Mowat-Wilson Syndrome

Area of Science:

• Computational biology
• Genetics
• Medical imaging

Background:

• Next-generation phenotyping (NGP) tools, including GestaltMatcher, have transformed rare genetic disorder diagnosis via computational facial analysis.
• While NGP is common in differential diagnosis, its use in variant reclassification within the ACMG framework is less explored.

Purpose of the Study:

• To investigate the utility of GestaltMatcher and multimodal data integration for variant reclassification in a patient with a suspected rare neurodevelopmental disorder.
• To establish Gestalt score thresholds for Phenotype-Phenotype-Genotype (PP4) evidence levels using Bayesian likelihood modeling.

Main Methods:

• Applied GestaltMatcher for facial analysis in a patient with a suspected Mowat-Wilson syndrome (MWS) and a de novo ZEB2 variant.
• Integrated Human Phenotype Ontology (HPO) terms and simulated exome data using the PEDIA framework for variant prioritization.
• Performed Bayesian likelihood modeling for Gestalt score thresholds and analyzed brain MRI for MWS-associated abnormalities.

Main Results:

• GestaltMatcher identified MWS as the top diagnosis; PEDIA integration confirmed ZEB2 as the likely causative gene.
• Facial images met moderate (3/4) and supporting (1/4) PP4 thresholds.
• Brain MRI showed corpus callosum thinning consistent with MWS; GestaltMatcher prioritized diagnosis in an infant case based on facial features alone.

Conclusions:

• NGP-driven facial phenotyping and multimodal integration show significant potential in dysmorphology.
• AI-assisted phenotyping can enhance diagnostic accuracy for neurodevelopmental disorders with distinctive facial features.