Skip to main content

Table 12 Performance results on Data-201706 focusing on the sub-ontology Organ abnormality

From: HPOAnnotator: improving large-scale prediction of HPO annotations by low-rank approximation with HPO semantic similarities and multiple PPI networks

MethodAUCAUPRmicro-AUCmicro-AUPRmacro-AUCmacro-AUPRleaf-AUCleaf-AUPR
NMF-Organ0.9550.5070.8830.2500.7450.1270.6820.077
NMF-PPN-Organ0.9620.5550.8890.2760.7550.1440.7010.091
NMF-NHPO-Organ0.9620.5350.8880.2640.7560.1410.7020.089
NMF-All0.9560.5120.8840.2580.7550.1290.6850.083
NMF-PPN-All0.9620.5530.8890.2730.7550.1430.6980.089
NMF-NHPO-All0.9620.5560.8890.2740.7550.1440.6990.090
HPOAnnotator-All0.9630.5590.8910.2780.7590.1460.7020.094
  1. The first three rows of methods with “Organ” are trained by HPO terms on Organ abnormality, while the others with “All” are trained by considering all sub-ontologies.
  2. Method performs best in terms of this evaluation metric are in boldface