Table 3.

Tokenization options for biomedical entity mentions [e.g. ‘brancio-oto-renal (BOR) syndrome’]

Option	Result
Tokens	‘branchio’, ‘oto’, ‘renal’, ‘BOR’, ‘syndrome’
MWE	‘branchio-oto-renal’, ‘(BOR)’, ‘syndrome’
Both	‘branchio-oto-renal’, ‘(BOR)’, ‘syndrome’, ‘branchio’, ‘oto’, ‘renal’, ‘BOR’, ‘syndrome’