Skip to main content

Table 4 The overall prediction accuracy achieved using Rotation Forest to each feature groups investigated in this study (in percentage).

From: Gram-positive and gram-negative subcellular localization using rotation forest and physicochemical-based features

Features

Gram-positive

Gram-negative

PSSM_AAC

75.7

71.2

PSSM_AC

79.5

72.2

OSD

63.9

63.9

OSA

67.7

68.9

PSSM_AAC + PSSM_AC

80.5

74.9

PSSM_AAC + PSSM_AC + OSD

81.3

75.6

PSSM_AAC + PSSM_AC + OSD + OSA

83.6

76.6

  1. It also shows the results achieved using Rotation Forest to the combination of feature groups to build the Combination of all for feature groups together. We use attribute number 9 (Hydrophobicity scale (contact energy derived from 3D data) attribute) for this experiment as using this attribute we attain our best results.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us