Skip to main content

Table 2 Average AUROC results (and standard deviations)

From: MOKPE: drug–target interaction prediction via manifold optimization based kernel preserving embedding

Methods

Data set

NR

GPCR

IC

E

Nearest profile

0.762* (0.013)

0.771* (0.004)

0.623* (0.010)

0.715* (0.012)

Weighted profile

0.768* (0.016)

0.813* (0.005)

0.765* (0.007)

0.783* (0.013)

LapRLS

0.760* (0.019)

0.810* (0.005)

0.752* (0.007)

0.774*(0.013)

DLapRLS

0.826 (0.015)

0.807* (0.011)

0.755* (0.012)

0.754* (0.011)

RLS-WNN

0.856 (0.015)

0.870 (0.006)

0.808 (0.009)

0.800* (0.012)

KBMF2K

0.798* (0.012)

0.810* (0.009)

0.792 (0.009)

0.724* (0.013)

CMF

0.806* (0.018)

0.807* (0.009)

0.767* (0.011)

0.795* (0.010)

WGRMF

0.874 (0.013)

0.878 (0.006)

0.801 (0.009)

0.822 (0.009)

GRGMF

0.874 (0.011)

0.879 (0.005)

0.814 (0.009)

0.825 (0.015)

HGBI

0.777* (0.013)

0.813* (0.004)

0.718* (0.008)

0.809 (0.014)

MOKPE

0.850 (0.015)

0.878 (0.005)

0.800 (0.007)

0.824 (0.010)

  1. * indicates that MOKPE significantly outperforms this method with \(p < 0.05\) using Mann–Whitney test. The highest result in each column is bolded and the second best is underlined
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us