Face2Gene is a suite of phenotyping applications that facilitate comprehensive and precise genetic evaluations. Face2Gene was developed by FDNA, the company, which is paving the way for clinicians to diagnose these diseases earlier, for labs to interpret genetic information more accurately and for life science researchers to make new discoveries to help develop rare disease therapies. As a confirmation, Face2Gene app increases the confidence of clinicians and bioinformatics by prioritizing genetic disorders and variants in the clinic and in the lab. This invention can detect phenotypes and reveal relevant facial and non-facial features.
FDNA was founded by Moti Shniberg and Prof. Lior Wolf. The company uses advanced technologies, including computer vision, deep learning, and artificial intelligence to analyze patient symptoms, features and genomic data in a network of thousands of genetics professionals worldwide, delivering scientific insights to improve and accelerate diagnostics and therapeutics, changing the lives of children with rare diseases.
Therefore, Face2Gene utilizes deep learning algorithms and computer vision in order to build syndrome-specific computational-based classifiers (syndrome gestalts). The proprietary technology converts a patient photo into de-identified mathematical facial descriptors (facial descriptors). The patient's facial descriptor is compared to syndrome gestalts to quantify similarity (gestalt scores) resulting in a prioritized list of syndromes with similar morphology. Artificial intelligence suggests likely phenotypic traits and genes assist in feature annotation and syndrome prioritization.
Face2Gene Connect™ gives developers the ability to access a patient's de-identified phenotypic data through a secure API. Adopting institutions can develop their own proprietary applications that leverage Face2Gene's unique gestalt analysis in combination with annotated clinical data, while streamlining and simplifying the workload for ordering clinicians.
The technology also provides laboratory teams with a detailed phenotype report to aid in the variant analysis by filtering and prioritizing variants based on clinical annotations of features, syndrome differentials and scoring of syndromes and variants using the proprietary feature and gestalt analysis.
The analysis has been shown to assist in whole exome sequencing interpretation. The deep phenotyping provided by Face2Gene analysis could be key in identifying syndromic forms of autism (lat. Autismus) and could help in the selection of diagnostic molecular tests. In addition, the detection rate for Cornelia de Lange syndrome was found to be comparable to dysmorphology experts.