Skip to main content

Table 4 Comparison of prediction results with different feature selection ratios on RPI1807

From: Predicting lncRNA–protein interactions through deep learning framework employing multiple features and random forest algorithm

Feature selection ratio

ACC (%)

SEN (%)

SPE (%)

MCC

b2s2_re

97.39

95.46

99.23

0.9734

b2s8_re

97.43

95.54

99.40

0.9738

b8s2_re

97.17

94.95

99.02

0.9711

b8s8_re

97.56

95.29

99.83

0.9750

  1. Bold values represent the maximum value of the corresponding evaluation indicator
  2. The number after b reflects the proportion of feature screening and retention for sequence, and the number after s represents the proportion of feature screening and retention for PPI, physicochemical properties and secondary structure
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us