The Gene Assay Database helps to achieve a genetic diagnosis for a pediatric population.
Client
Genome Medicine Center & Research Informatics, Children’s Mercy Hospital
Technology used
React, .NET, Cosmos DB, Azure App Service
Situation
Dr. Joe Alaimo, Assistant Director of Molecular Genetics at Children’s Mercy Hospital, saw a critical need for a tool that provides a broader and more accessible variety of information on disease-causing DNA variants in a pediatric population. Despite great strides in sequencing technology in recent years, 70% – 90% of genetic illnesses in children remain undiagnosed. Two factors heavily influence the inability to diagnose:
1. insufficient data on the frequency and impact of these variants
2. variability in interpretation across genetic diagnostic laboratories.
Dr. Alaimo envisioned a central database that would display summaries, structural information (visual depiction), and functional information for variants on any given gene. This tool should enable well-informed determinations about the effects of specific variations on children, and ultimately lead to broader diagnostic coverage for children with genetic illnesses.
Challenge
Our overarching challenge throughout this project was to develop a strong enough understanding of the current gaps in available genetic diagnostic tools to empower us to architect a cost effective and intuitive way to present gene variant information to diagnosticians.
- We needed a data set that enables users to easily search for their gene of interest.
- We needed dynamic visual representation of known variant locations on the gene, with the ability to filter the variant display by variant classification from the most recent ACG clinical guidelines.
- We needed to display functional information for variants on the gene, with the ability for users to search by protein or by nucleotide and have ready access to additional information about the variant.
Solution
We started the project kickoff by documenting and exploring all the tools and applications used by a Clinical Geneticist and outlined the typical workflow within the lab. We needed to fully understand the context to know where to avoid duplication of efforts – needs that might be met elsewhere. With this knowledge base, we documented the abilities of the user types – Administrators, who would need to have the ability to curate and updating meta-data about variants, and Readers, users who will access the tool on the web.
- We seeded our database with a complete set of genes with known variants from ClinVar, stored in a non-relational database with an elastic search, allowing users to quickly land on their gene of interest.
- We mapped variant locations onto a canvas using custom CSS, generating a graphic for each gene that shows the locations of protein domains and variants.
- We created editable lists of assays and known outcomes associated with variants, accompanied by links to relevant PubMed articles.
Key Benefits
- Within a lab, reduces time spent towards diagnosis by 70%
- Enables institutions to collect meaningful variant insights in order to share meaningful outcomes and observations within their clinical population
- Strengthens diagnostic success and quickens the path to treatment for genetic illnesses
