Fabric Genomics and Rady Children's Institute for Genomic Medicine® today announced the publication of a retrospective study in Genome Medicine showing that across six leading genomic centers and hospitals, researchers were able to detect more than 90% of disease-causing variants in infants with rare diseases using the Fabric GEM AI algorithm and whole-genome and whole-exome data from previously diagnosed newborns and rare disease patients at Rady Children's Hospital - San Diego and other clinical sites.
Despite differences in case collection, sequencing methods, and bioinformatics pipelines across all sites, Fabric GEM's performance demonstrated a new standard of accuracy, ranking the causative variant first or second more than 90% of the time. In addition, Fabric GEM ranked specific diseases and conditions associated with these genes to assist clinicians in the ultimate diagnosis of each case. These findings demonstrate how artificial intelligence (AI) can successfully reduce the burden of gene variant review by clinical geneticists.
Fast and definitive genetic diagnosis is essential to providing the right treatment in a timely manner for critically ill newborns. Fabric GEM has successfully demonstrated that it can automatically and quickly suggest a very short list of candidate genes for interpretation through whole-genome or whole-exome sequencing."
Stephen Kingsmore, MD, DSc, Study Co-Author, President and CEO, Rady Children's Institute for Genomic Medicine
Additional centers that participated in the study include the University of Utah, Boston Children's Hospital, Christian-Albrechts University of Kiel & University Hospital Schleswig-Holstein, HudsonAlpha Institute of Biotechnology, Tartu University Hospital, and the Translational Genomics Research Institute (TGen).
"This study is an exciting milestone demonstrating how AI-powered decision support technologies can empower clinicians. It has the potential to significantly improve patient care with rapid insights distilled from clinical notes, medical databases, and genome sequences. Human review of these critical, but ever-expanding data is becoming infeasible due to their size and complexity. Hence GEM," said Mark Yandell, PhD, Professor of Human Genetics and Edna Benning Presidential Endowed Chair at the University of Utah, a founding scientific advisor to Fabric and a co-author on the paper.
This study also demonstrated the use of Clinithink's CLiX focus, a natural language processing (NLP) technology applied to medical notes recorded in electronic medical records. When compared to manual abstraction, this automated approach, which couples Clinithink's NLP technology with the Human Phenotype Ontology and Fabric GEM, can rival the results achieved through time-intensive, expert-driven curation.
"Finally, clinicians do not have to sacrifice accuracy for speed when faced with a possible rare disease diagnosis in a critical setting like the NICU where time is of the essence," said Martin Reese, PhD, CEO of Fabric Genomics and a co-author on the paper. "This study provides the rigorous benchmark validation required for its use in the clinic, showcasing how any hospital can bring informed genomics to their patients."
De La Vega, F.M., et al. (2021) Artificial intelligence enables comprehensive genome interpretation and nomination of candidate diagnoses for rare genetic diseases. Genome Medicine. doi.org/10.1186/s13073-021-00965-0.