Skip to main content

Table 1 Performance of top esRF models generated in this study

From: Prediction of unconventional protein secretion by exosomes

 

Dataset translation

PCP

AA

PCP + AA

DP

PCP + AA + DP

I

4500

3500

3000

3000

5000

ACC (%)

64.74 ± 1.94

68.61 ± 2.05

67.94 ± 2.02

69.88 ± 2.08

69.74 ± 1.93

AUC

0.69 ± 0.04

0.75 ± 0.03

0.72 ± 0.06

0.76 ± 0.03

0.76 ± 0.04

MCC

0.29 ± 0.06

0.37 ± 0.04

0.36 ± 0.05

0.40 ± 0.05

0.39 ± 0.05

  1. Table reports the accuracy (ACC), area under the curve (AUC) and Matthews correlation coefficient (MCC) of the top performing esRF models obtained with the indicated interaction value (I) by training in physico-chemical properties (PCP), amino acid composition (AA), combination of physico-chemical properties and amino acid composition (PCP + AA), dipeptide composition (DP) and combination of physico-chemical properties, amino acid composition and dipeptide composition (PCP + AA + DP). Performance values were obtained under 10-cross validation experiments that were repeated 10 times