Last week I talked about the 100,000 Genomes Project in the UK. That video tip was an introduction and overview of the project. This week, though, we’ll highlight one software piece that is in place now. PanelApp is a collection of gene lists, with information about the evidence associated with variations in these genes. The goal is to inform clinical interpretation based on the best quality evidence.
It’s a straigtforward interface, with a traffic-light based system (green, yellow, red) to assist with quick diagnostic assessments. The lists are based on a set of rare disease categories and phenotypes, and the idea is that qualified reviewers will create a definitive list of evidence for certain variations, and rate the genes accordingly. Ultimately they want to create a consensus gene panel collection. They describe the evidence thresholds this way:
- Green = highest level of confidence; a gene from 3 or 4 sources.
- Amber = intermediate; a gene from 2 sources.
- Red = lowest level of confidence; 1 of the 4 sources or from an expert list.
You can learn more about the strategy from their announcement: New rare disease gene tool launched – PanelApp. This first video will give you an idea of how this looks. A public interface is available for anyone to look around and download the information. You don’t need to log in, you can click on the “Browse Panels” tab for an idea of the organization and information content.
So crowd-sourcing this evidence is the idea here. And they are interested in having people contribute to these panels. A separate video describes the roles for people who are interested in contributing to the collection and curation of the information. Reviewers will evaluate the existing items and also add new genes. They have detailed instructions and guidelines available to help. However, the first round of review ended already (October 19 2015). The panels will remain available to explore and download. But there will be another review period opening in the future. So if you are considering a contribution to this work, not to worry–your chance will come and it won’t hurt to be prepared. Also, if you are considering using the panels, I think it’s a good idea to know how the curation and review is done.
The information in the panels now has been seeded by many key sources that you may be familiar with. They have used data from OMIM, OrphaNet, Genetics Home Reference, the Decipher DDD project, HGNC, ClinVar, UniProt, NCBI Nucleotide, and ENSEMBL among others to add phenotype and gene details of many types. So for any gene you can keep digging for addtional information.
There are a long more details over at the announcment post: New rare disease gene tool launched – PanelApp. But I think the structure and color cues are helpful. (Wait, colour, right? UK variation). Check it out yourself, though, with the link below. You can hear more in this podcast as well.
— Daniel S. Levine (@dslevine) October 19, 2015
100,000 Genomes Project: http://www.genomicsengland.co.uk/
Siva, N. (2015). UK gears up to decode 100 000 genomes from NHS patients The Lancet, 385 (9963), 103-104 DOI: 10.1016/S0140-6736(14)62453-3