data.frame of 46 rows, with 2 columns:
rankid - a numeric rank id, consecutive
ranks - a comma separated vector of names that are considered equal to one another within the row
Details
We use this data.frame to do data sorting/filtering based on the ordering of ranks.
Please let us know if there is a rank that occurs from one of the data
sources taxize that we don't have in rank_ref
dataset.
Let us know if you disagree with the ordering of ranks.
Note that rankid
280 are essentially "genetic variants"; placed just above
'unspecified' to denote they're not without rank, but they're not
really taxonomic ranks either. As far as I know there's no way
to delineate among these "genetic variant" types.