Table 1.

Summary of datasets.

Data Source	Recon 2	Natural Products	Marketed Drugs	FLUOROPHORES	ZINC15-subset1	ZINC15-subset2	Total
Original smiles count	1112	158,809	1381	143	249,455	6,202,415	6,613,315
Canonical smiles count	1056	148,971	1357	115	247,804	4,425,891	4,825,194 (4,774,258-unique)