After nearly a decade (!) of iterative development on the seqr platform for management and analysis of rare disease genomic data, we have a preprint up describing this critical piece of our rare disease infrastructure:
Conversation
seqr has been our workhorse platform for collaborations with groups around the world through the Center for Mendelian Genomics; to date it's been used to study more than 10,000 families, to make more than 3,800 diagnoses, and find >300 new disease genes.
1
3
It's been critical for our approach to rare disease research, because (1) it scales to massive numbers of families with WGS data, (2) it's accessible to any collaborator through a web browser, and (3) it facilitates collaboration and data sharing.
1
2
seqr has always been open-source, and is under very active development (e.g. recently added support for structural variants and RNA-seq data). There are now deployments in Australia, Bahrain, Estonia, UK, and multiple US labs. Code here:
Replying to
At seqr's heart is the analysis interface, which makes it easy to filter variants by inheritance, frequency, functional impact, known pathogenicity, and quality; review read data and external evidence; and tag variants with shared notes.
1
1
12
Lots more information in the preprint - huge thanks to and Hana Snow for tireless work on the platform; to our analysts and users; and especially to and for many contributions to the seqr workflow and for driving the paper!
6
