Sequence.quality: Quality metrics (such as truthTP, queryTP) indicate the overall analytic accuracy of variant calls in a region, but don't provide the analytic accuracy for a specific base call. How are the existing fields used clinically, and is there value in providing additional attributes to convey the accuracy of a specific base call? (Would it make sense for us to begin to define Quality use cases, and from there revisit what we need in Sequence resource?)