Artificial neural network models for prediction of intestinal permeability of oligopeptides

Abstract

Background: Oral delivery is a highly desirable property for candidate drugs under development. Computational modeling could provide a quick and inexpensive way to assess the intestinal permeability of a molecule. Although there have been several studies aimed at predicting the intestinal absorption of chemical compounds, there have been no attempts to predict intestinal permeability on the basis of peptide sequence information. To develop models for predicting the intestinal permeability of peptides, we adopted an artificial neural network as a machine-learning algorithm. The positive control data consisted of intestinal barrier-permeable peptides obtained by the peroral phage display technique, and the negative control data were prepared from random sequences.

Results: The capacity of our models to make appropriate predictions was validated by statistical indicators including sensitivity, specificity, enrichment curve, and the area under the receiver operating characteristic (ROC) curve (the ROC score). The training and test set statistics indicated that our models were of strikingly good quality and could discriminate between permeable and random sequences with a high level of confidence.

Conclusion: We developed artificial neural network models to predict the intestinal permeabilities of oligopeptides on the basis of peptide sequence information. Both binary and VHSE (principal components score Vectors of Hydrophobic, Steric and Electronic properties) descriptors produced statistically significant training models; the models with simple neural network architectures showed slightly greater predictive power than those with complex ones. We anticipate that our models will be applicable to the selection of intestinal barrier-permeable peptides for generating peptide drugs or peptidomimetics.

Figure 1

Predictive features of the model. The model was constructed with zero neuron in a hidden layer and one in an output layer using binary descriptors. (A) Enrichment curve, (B) Histogram Actives vs. Model values, and (C) Receiver Operating Characteristic (ROC) curve. The features for the training and test set were plotted in the left and right panels, respectively.

Figure 2

Distribution of prediction scores for all permutations of three peptide sequences.

Figure 3

The features of the model constructed with the decoy set. The models were constructed with zero neuron in a hidden layer and one in an output layer using binary descriptor. (A) Training set and (B) Test set.

Figure 4

A schematic view of peroral phage display procedure. After the third round of biopanning, individual recombinant phage was randomly selected from each organ tissue elute for analysis of peptide sequences from their genomes.