Using PPI network autocorrelation in hierarchical multi-label classification trees for gene function prediction

Stojanova, Daniela; Ceci, Michelangelo; Malerba, Donato; Dzeroski, Saso

doi:10.1186/1471-2105-14-285

BMC Bioinformatics

Table 2 The performance of NHMC and competitive methods in predicting gene function for different datasets and PPI networks

From: Using PPI network autocorrelation in hierarchical multi-label classification trees for gene function prediction

	All genes
Network		DIP				BioGRID
Method	CLUS-HMC	NHMC		FF	H	NHMC		FF	H
Dataset		α = 0	α = 0.5			α = 0	α = 0.5
seq	0.023	0.032	0.030	0.004	0.003	0.011	0.011	0.006	0.006
pheno	0.019	0.016	0.016	0.001	0.001	0.016	0.016	0.003	0.002
struc	0.018	0.012	0.012	0.001	0.001	0.012	0.012	0.003	0.002
homo	0.040	0.013	0.013	0.000	0.000	0.012	0.012	0.001	0.002
cellcycle	0.019	0.287	0.288	0.004	0.003	0.012	0.012	0.006	0.006
church	0.014	0.015	0.012	0.003	0.002	0.012	0.012	0.006	0.006
derisi	0.017	0.015	0.017	0.004	0.003	0.044	0.317	0.006	0.006
eisen	0.030	0.024	0.024	0.005	0.003	0.015	0.334	0.006	0.008
gasch1	0.024	0.018	0.019	0.003	0.002	0.050	0.354	0.006	0.006
gasch2	0.020	0.021	0.021	0.004	0.003	0.012	0.012	0.006	0.006
spo	0.019	0.018	0.015	0.004	0.003	0.012	0.012	0.006	0.006
exp	0.023	0.017	0.016	0.003	0.002	0.012	0.012	0.006	0.006
Average:	0.022	0.041	0.040	0.003	0.002	0.018	0.093	0.005	0.005
	Highly connected genes
Network		DIP				BioGRID
Method	CLUS-HMC	NHMC		FF	H	NHMC		FF	H
Dataset		α = 0	α = 0.5			α = 0	α = 0.5
seq	0.037	0.072	0.1	0.003	0.001	0.025	0.035	0.007	0.007
pheno	0.051	0.016	0.051	0.002	0.002	0.051	0.051	0.006	0.005
struc	0.078	0.078	0.078	0.001	0.002	0.078	0.078	0.003	0.003
homo	0.047	0.068	0.068	0.001	0.001	0.023	0.023	0.002	0.003
cellcycle	0.027	0.036	0.018	0.004	0.005	0.026	0.041	0.007	0.007
church	0.017	0.025	0.025	0.004	0.004	0.025	0.025	0.007	0.007
derisi	0.078	0.078	0.106	0.004	0.004	0.044	0.042	0.007	0.007
eisen	0.043	0.061	0.146	0.005	0.005	0.030	0.045	0.007	0.007
gasch1	0.051	0.094	0.095	0.004	0.005	0.050	0.046	0.007	0.007
gasch2	0.04	0.088	0.107	0.004	0.005	0.025	0.043	0.007	0.007
spo	0.04	0.078	0.09	0.004	0.005	0.026	0.035	0.007	0.007
exp	0.045	0.036	0.092	0.004	0.004	0.025	0.025	0.007	0.007
Average:	0.046	0.061	0.081	0.003	0.003	0.036	0.041	0.006	0.006

We use the 2/3-1/3 training-testing evaluation schema. We report the $\bar{AUPRC}$ of the CLUS-HMC (α = 1), NHMC (α = 0.5 and α = 0), FunctionalFlow (FF), and Hopfield (H) methods, when predicting gene function in yeast using GO annotations. We use 12 yeast (Saccharomyces cerevisiae) datasets (as considered by [1]). We consider all genes. Results for two PPI networks (DIP and BioGRID) are presented.

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com