Implementing Phenopackets in a Variant Discovery Pipeline

Variant filtration is key to rare disease discovery. For a rare disease discovery app to be successful, it must be able to cleanly represent the relevant variants in a patient’s genome. Phenopackets present a method of filtering genes based on a patient’s phenotype. Phenotype data can narrow down the set of at risk genes clinicians need to parse during rare disease discovery using tools like exomiser. This project will use phenopackets to dynamically filter variants for the purpose of rare disease discovery. This project has two deliverables, each with their own set of requirements: A method to filter genes using phenopacket data. Must be optimized for speed, given this is a dynamic application. However, speed will be limited by the FHIR Genomics Operations reference implementation response time. A clean, easily understandable presentation of variants useful for manual review. Two views should be provided: one for the most at risk variants, and another for a more detailed view of each variant in at risk genes. The former should have a graphical component. The latter does not need to have as clean of a presentation. Users should be able to identify the most at risk variants within 5-10 seconds of the page loading.