Skip to main content

Table 3 The results of validation for models with network architecture (7 × 20)-0-1a.

From: Artificial neural network models for prediction of intestinal permeability of oligopeptides

Leave-5%-out cross-validationb

Decoy analysisc

1 : 1 Data set

Real set

Decoy set

Trainingd

Testd

Training

Test

Training

Test

0.841 ± 0.002

0.760 ± 0.005

0.82

0.74

0.70

0.47

  1. a The network architecture (7 × 20)-0-1 specifies a model constructed with zero neuron in hidden layer and one in output layer using the binary descriptor.
  2. b The results of rigorous test using leave-5%-out method in 1:1 data sets.
  3. c Comparison of ROC scores between real and decoy set using non-redundant data.
  4. d The results of 20 rigorous tests are averaged and expressed as mean ± standard deviation.
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us