Fabric Genomics, Inc., today announced that the Broad Institute of MIT and Harvard has selected the company to support the development and implementation of its clinical whole genome sequencing-based offerings. As part of the agreement, Fabric Genomics will provide the AI-based clinical decision support platform used to support the interpretation of clinical whole genomes and generate patient reports. The collaborators will launch the clinical whole genome interpretation service as part of clinical utility studies with a national healthcare system.
Fabric’s AI-powered clinical decision support platform includes Fabric ACE and Fabric GEM, two proprietary technologies that support highly accurate variant detection while providing measurable time savings and significantly lower analysis costs. The clinical analysis services are powered by a world-class team of variant interpretation geneticists that have interpreted over 10,000 clinical cases in 2021 alone.
“Many people with rare diseases, including newborns, do not have easy access to this type of high complexity clinical whole genome testing which can have a profound impact on health outcomes,” said Martin Reese, PhD, Co-Founder and CEO of Fabric Genomics. “We are very proud to support the Broad Institute in providing healthcare providers and patients greater access to genomics-driven precision medicine.”
Fabric’s clinical genomics decision-support system for whole genomes, recently published in the New England Journal of Medicine as part of Genomics England’s 100,000 Genomes Project on rare disease diagnosis, uses artificial intelligence to automate analyses of sequencing data, including single nucleotide variants alongside complex structural variants, which has traditionally been a challenge. It utilizes a patient’s clinical information and probabilistic disease matching to identify not only candidate genes but also to prioritize diagnoses. The data sources and logic utilized by the Fabric GEM algorithm are fully transparent in the software, resulting in high diagnostic confidence. The resulting reports allow clinical teams to concentrate on the most likely possibilities, reducing the time to a genetic diagnosis from days to minutes. Fabric’s GEM algorithm has been shown in a growing number of peer-reviewed studies to detect more than 90% of disease-causing variants as the top two candidates.
Fabric ACE takes sequencing data as an input and uses a forward-chaining inference engine to inform the classification of variants in clinically relevant genes according to criteria established by the American College of Medical Genetics and Association for Molecular Pathology. Ultimately, this allows for more accurate and faster clinical reporting allowing genomic laboratories to scale their testing.
