It varies but for these three datasets, I think we mostly consume all samples that passed quality metrics and were not related. The exception is that I think there was a subset of ESP that did not have sufficient consent (perhaps a quarter?? - I’d need to investigate further) that we were not allowed to include. Also, for TOPMed, we only have the data from those samples sequenced at the Broad which is about 30,000 so a much smaller fraction than the total TOPMed dataset. If you have specific questions, I’d ask Sam Baxter at samantha@broadinstitute.org
1 Like