Skip to main content

Table 1 Details of the data sets used for evaluating the label similarity-incorporated classifiers using a cross-validation methodology.

From: Incorporating functional inter-relationships into protein function prediction algorithms

Dataset # Proteins # Features # Classes k
Mnaimneh 4062 215 137 20
Rosetta 3980 300 137 20
Krogan 2117 2117 108 5
Combined 3762 4277 136 10