Few shot learning for phenotype-driven diagnosis of patients with rare genetic diseases

Abstract

There are over 7,000 rare diseases, some affecting 3,500 or fewer patients in the US. Due to clinicians limited experience with such diseases and the heterogeneity of clinical presentations, approximately 70% of individuals seeking a diagnosis remain undiagnosed. Deep learning has demonstrated success in aiding the diagnosis of common diseases. However, existing approaches require labeled datasets with thousands of diagnosed patients per disease. We present SHEPHERD, a few shot learning approach for multi-faceted rare disease diagnosis. SHEPHERD performs deep learning over a knowledge graph enriched with rare disease information and is trained primarily on simulated rare disease patients. We demonstrate SHEPHERDs effectiveness across diverse diagnostic tasks, performing causal gene discovery, retrieving "patients-like-me", and characterizing novel disease presentations, using real-world cohorts from the Undiagnosed Diseases Network (N = 465), MyGene2 (N = 146), and the Deciphering Developmental Disorders Study (N = 1, 431). SHEPHERD demonstrates the potential of knowledge-guided deep learning to accelerate rare disease diagnosis.