I am working on automatic diagnosis of monogenic diseases using machine learning and natural language processing techniques. Monogenic diseases are caused by mutations to a single gene in the human genome. Because human genome sequencing has become comparatively inexpensive in recent years, our knowledge of genetic mutations that cause genetic diseases is growing rapidly. This knowledge is deposited in hundreds of thousands of research articles. I am working on a system that takes information about a patient’s symptoms, as well as their genotype, and discovers diagnoses for patients with monogenic diseases by automatically reading the relevant literature. Because diagnosis of monogenic diseases is so difficult and our knowledge is still incomplete, patients and their relatives sometimes wait for years to receive a definitive diagnosis, which exacts a psychological toll on them and hinders effective treatment. My project aims to speed up the costly and slow process of diagnosis and make it accessible to a broader audience.