Fig. 2.

The results of swapping and deassigning stereospecific assigned DRs. Most entries have none or only small percentages of these modifications indicating that the stereospecific information was treated correctly. A few entries required swapping of all stereo groups as indicative of a nomenclature problem. Floating chirality was used in another set of entries. Both issues are resolved by this procedure. The x-axis has been truncated to showcase the 1,630 entries with percentages above zero, out of a total set of 4,588 entries with converted DR