Table 1. Area under the curve (AUC) for the four different corpora, with and without document weight for scoring co-occurrences.

	Disease–gene		Protein–protein		Protein–compartment
	With document weight	Without document weight	With document weight	Without document weight	With document weight	Without document weight
All full-texts	0.91	0.86	0.73	0.70	0.71	0.67
Core full-texts	0.90	0.86	0.73	0.70	0.71	0.67
Core abstracts	0.81	0.82	0.70	0.70	0.62	0.64
MEDLINE	0.85	0.85	0.70	0.71	0.65	0.66