Further delineation of Ayme-Gripp syndrome and use of automated facial analysis tool

Ayme-Gripp syndrome (AGS) is an autosomal dominant multisystem disorder caused by specific heterozygous variants in MAF. The resulting aberrant protein shows impaired GSK-mediated MAF phosphorylation. AGS is characterized by congenital cataracts, sensorineural hearing loss, short stature, intellectual disability, and distinctive facial features with brachycephaly. Cardiac and joint phenotypes are present in nearly half of patients. We review information on 10 published individuals with MAF mutations and clinical AGS and describe five additional patients, including three with novel mutations. Joint problems, typically including radioulnar synostosis and joint limitations, were present in 9/15 patients. Hip replacement in young adulthood was needed in four patients. Pericarditis occurred in 6/15 individuals. An automated facial analysis of 2D photos was used to compare the facial phenotype of 13 individuals from the literature or reported here, with facial photos of a control cohort of unaffected individuals and a cohort of Down syndrome patients. A multiclass approach yielded an accuracy of 86.86% and 89.05%, respectively, in two independent experiments compared to a random chance of 37.74%. In binary comparisons of AGS and Down syndrome, the area under the curve (AUC) was 0.994 (P<.001) and 1.0 (P<.001), respectively. Binary comparisons of AGS and unaffected controls yielded AUC of 0.994 (P<.001) and 0.989 (P=.003), respectively, suggesting that the facial phenotype of AGS could clearly be distinguished from unaffected individuals and from Down syndrome patients. Automated facial analysis may be helpful in the identification and evaluation of individuals suspected to have AGS.