Table 2

Statistics comparing BioCreative II gene normalization task (human) with BioCreative I tasks (mouse, fly, and yeast).


Number of unique IDs
Average synonym length in words
Average synonyms per identifier
Average identifiers per synonym (ambiguity)
BioCreative maximum recall @ precision
BioCreative maximum F-measure

Human
32,975
2.17
5.55
1.12
0.88 @ 0.50
0.81
Mouse
52,494
2.77
2.48
1.02
0.90 @ 0.43
0.79
Yeast
7,928
1.00
1.86
1.01
0.96 @ 0.65
0.92
Fly
27,749
1.47
2.94
1.09
0.84 @ 0.73
0.82

Statistics on synonyms are based on lexical resources provided by the task organizers.

Morgan et al. Genome Biology 2008 9(Suppl 2):S3   doi:10.1186/gb-2008-9-s2-s3

Open Data