Table 1.

Comparison of annotation counts between tokenization approaches

Re-tokenization	Protein	Cellular	Tissue	Molecule	Cell	Organism
Without	97.178	99.772	95.951	96.107	97.099	93.691
With	99.187	99.866	99.842	99.424	99.559	98.921

The comparison of annotation counts between preprocessing with only NERsuite tokenization module (without) and with both NERsuite tokenization and additional tokenization (with). The numbers are percents of annotations compared to the provided data presented for each entity type.