From face to gene

In a multi-center study, we are currently investigating the diagnostic performance of the analysis tool Face2Gene in assessing exome data (PEDIA-study,

Especially, in patients with GPI anchor deficiencies that show a characteristic facial gestalt a correct prediction of the affected gene can be estimated with high accuracy. To improve accuracy and precision we are constantly looking for diagnosed or suspected patients with GPI biosynthesis defects (GPIBDs).

In this project we are using a frontal image of the face, clinical data (in HPO terms), and the underlying mutation to train a classifier for the prediction of the affected gene. All data is collected anonymously, the facial image will not be shared or publicly available, only biometric data will be collected.

Using this technology we are improving the diagnostic procedure of novel cases and optimizing the care of existing patients.

Clinicians as well as parents are encouraged to contact us for further details.

All information including a consent form for the facial picture can be send to: alexej.knaus(at)