Skip to main content

Table 9 Compared with existing methods on PDB1075 data set (LOOCV)

From: A sequence-based multiple kernel model for identifying DNA-binding proteins

Methods ACC (%) MCC SN (%) Spec (%)
IDNA-Prot|dis 77.30 0.54 79.40 75.27
PseDNA-Pro 76.55 0.53 79.61 73.63
IDNA-Prot 75.40 0.50 83.81 64.73
DNA-Prot 72.55 0.44 82.67 59.76
DNAbinder 73.95 0.48 68.57 79.09
iDNAPro-PseAAC 76.56 0.53 75.62 77.45
Kmer1+ACC 75.23 0.50 76.76 73.76
Local-DPP 79.10 0.59 84.80 73.60
MKSVM-HKA 81.30 0.63 82.29 80.36
MSFBinder 83.35 0.67 83.62 83.09
FKRR-MVSF 83.26 0.67 85.71 80.91
Our method (MKSVM with MKL-CKA) 84.19 0.68 85.91 82.55
  1. The bold font indicates the largest value in the column