Skip to main content
Figure 7 | BMC Bioinformatics

Figure 7

From: Cancer module genes ranking using kernelized score functions

Figure 7

Integrated network obtained by combining FI and HumanNet networks: comparison of precision and recall curves between our proposed kernelized score functions and other machine learning methods for gene ranking. precisions, averaged across the 298 Cancer Modules, are computed through 5-fold cross-validation techniques repeated 5 times for different fixed recall levels ranging from 0.1 to 1. S AV (Average score), S NN (Nearest-neighbor score) and S kNN (k-Nearest-neighbor score) represent kernelized score functions. The parameter k of S kNN is set to 27. Zhou is the algorithm based on Gaussian Random Fields proposed in [31] and GeneMANIA its variant, while LabelProp is the Label Propagation algorithm proposed in [17].

Back to article page