Skip to main content
Fig. 6 | BMC Bioinformatics

Fig. 6

From: GODoc: high-throughput protein function prediction using novel k-nearest-neighbor and voting algorithms

Fig. 6

Fmax (a) and precision-recall curve (b) of each method on CAFA2-benchmark trained on CAFA2-Swiss, where Fixed, Dyn.Inverse, Dyn.FunOverlap, and Hybrid represent Fixed-KNN, Dynamic-KNN with Inverse voting weight, Dynamic-KNN with FunOverlap voting weight, and Hybrid-KNN, respectively. The number inside the bar shows the predicted proportion, in particular for Dynamic-KNN in partial mode

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us