bobular
commented
1 year ago This is for weighting, for example, male/female donut markers by the sample.specimen_count?
Would this be related to https://github.com/VEuPathDB/EdaNewIssues/issues/628 ?
Maybe both would be best handled by the "abundance marker mode" that we've vaguely discussed?
ignore the variable variable vocabulary problem here. we only count what exists.
leave the door open though, in case we get into no data debates in the sam in the future
immediately drop/ ignore from the table/ stream anything we filtered against.
use the N annotated variable to keep a sum of counts per value of the user-selected variable.
i believe this should expand easily to accommodate proportions as well as counts