Table 1.

The statistic of our gene corpus.

Data set	Articles	Gene mentions (gene/family/domains)	Gene identifiers
BioCreative II GN training set	281	3,019/1,115/278	758
BioCreative II GN test set	262	3,233/1,252/361	928
NLM Citation GIA test collection	151	1,205/160/17	310
Total	694	7,457/2,527/656	1996